Upload 12 files

.gitattributes

@@ -25,7 +25,6 @@
 *.safetensors filter=lfs diff=lfs merge=lfs -text
 saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
 *.tflite filter=lfs diff=lfs merge=lfs -text
 *.tgz filter=lfs diff=lfs merge=lfs -text
 *.wasm filter=lfs diff=lfs merge=lfs -text
@@ -33,3 +32,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.index filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.jpg filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,3 +1,14 @@
 ---
-license: apache-2.0
+title: Genshin Impact RVC Models (combined)
+emoji: 🎤
+colorFrom: purple
+colorTo: red
+sdk: gradio
+sdk_version: 3.36.1
+app_file: app.py
+pinned: false
+license: mit
+duplicated_from: taishoushiki/combined-GI-RVC-models
 ---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app-full.py

@@ -0,0 +1,503 @@
+import os
+import glob
+import json
+import traceback
+import logging
+import gradio as gr
+import numpy as np
+import librosa
+import torch
+import asyncio
+import edge_tts
+import yt_dlp
+import ffmpeg
+import subprocess
+import sys
+import io
+import wave
+from datetime import datetime
+from fairseq import checkpoint_utils
+from lib.infer_pack.models import (
+    SynthesizerTrnMs256NSFsid,
+    SynthesizerTrnMs256NSFsid_nono,
+    SynthesizerTrnMs768NSFsid,
+    SynthesizerTrnMs768NSFsid_nono,
+)
+from vc_infer_pipeline import VC
+from config import Config
+config = Config()
+logging.getLogger("numba").setLevel(logging.WARNING)
+limitation = os.getenv("SYSTEM") == "spaces"
+
+audio_mode = []
+f0method_mode = []
+f0method_info = ""
+
+if limitation is True:
+    audio_mode = ["Upload audio", "TTS Audio"]
+    f0method_mode = ["pm", "harvest"]
+    f0method_info = "PM is fast, Harvest is good but extremely slow, Rvmpe is alternative to harvest (might be better). (Default: PM)"
+else:
+    audio_mode = ["Input path", "Upload audio", "Youtube", "TTS Audio"]
+    f0method_mode = ["pm", "harvest", "crepe"]
+    f0method_info = "PM is fast, Harvest is good but extremely slow, Rvmpe is alternative to harvest (might be better), and Crepe effect is good but requires GPU (Default: PM)"
+
+if os.path.isfile("rmvpe.pt"):
+    f0method_mode.insert(2, "rmvpe")
+
+def create_vc_fn(model_name, tgt_sr, net_g, vc, if_f0, version, file_index):
+    def vc_fn(
+        vc_audio_mode,
+        vc_input, 
+        vc_upload,
+        tts_text,
+        tts_voice,
+        f0_up_key,
+        f0_method,
+        index_rate,
+        filter_radius,
+        resample_sr,
+        rms_mix_rate,
+        protect,
+    ):
+        try:
+            print(f"Converting using {model_name}...")
+            if vc_audio_mode == "Input path" or "Youtube" and vc_input != "":
+                audio, sr = librosa.load(vc_input, sr=16000, mono=True)
+            elif vc_audio_mode == "Upload audio":
+                if vc_upload is None:
+                    return "You need to upload an audio", None
+                sampling_rate, audio = vc_upload
+                duration = audio.shape[0] / sampling_rate
+                if duration > 20 and limitation:
+                    return "Please upload an audio file that is less than 20 seconds. If you need to generate a longer audio file, please use Colab.", None
+                audio = (audio / np.iinfo(audio.dtype).max).astype(np.float32)
+                if len(audio.shape) > 1:
+                    audio = librosa.to_mono(audio.transpose(1, 0))
+                if sampling_rate != 16000:
+                    audio = librosa.resample(audio, orig_sr=sampling_rate, target_sr=16000)
+            elif vc_audio_mode == "TTS Audio":
+                if len(tts_text) > 100 and limitation:
+                    return "Text is too long", None
+                if tts_text is None or tts_voice is None:
+                    return "You need to enter text and select a voice", None
+                asyncio.run(edge_tts.Communicate(tts_text, "-".join(tts_voice.split('-')[:-1])).save("tts.mp3"))
+                audio, sr = librosa.load("tts.mp3", sr=16000, mono=True)
+                vc_input = "tts.mp3"
+            times = [0, 0, 0]
+            f0_up_key = int(f0_up_key)
+            audio_opt = vc.pipeline(
+                hubert_model,
+                net_g,
+                0,
+                audio,
+                vc_input,
+                times,
+                f0_up_key,
+                f0_method,
+                file_index,
+                # file_big_npy,
+                index_rate,
+                if_f0,
+                filter_radius,
+                tgt_sr,
+                resample_sr,
+                rms_mix_rate,
+                version,
+                protect,
+                f0_file=None,
+            )
+            info = f"[{datetime.now().strftime('%Y-%m-%d %H:%M')}]: npy: {times[0]}, f0: {times[1]}s, infer: {times[2]}s"
+            print(f"{model_name} | {info}")
+            return info, (tgt_sr, audio_opt)
+        except:
+            info = traceback.format_exc()
+            print(info)
+            return info, None
+    return vc_fn
+
+def load_model():
+    models = []
+    with open(f"weights/model_info.json", "r", encoding="utf-8") as f:
+        models_info = json.load(f)
+    for character_name, info in models_info.items():
+        if not info['enable']:
+            continue
+        model_title = info['title']
+        model_name = info['model_path']
+        model_author = info.get("author", None)
+        model_cover = f"weights/{character_name}/{info['cover']}"
+        model_index = f"weights/{character_name}/{info['feature_retrieval_library']}"
+        cpt = torch.load(f"weights/{character_name}/{model_name}", map_location="cpu")
+        tgt_sr = cpt["config"][-1]
+        cpt["config"][-3] = cpt["weight"]["emb_g.weight"].shape[0]  # n_spk
+        if_f0 = cpt.get("f0", 1)
+        version = cpt.get("version", "v1")
+        if version == "v1":
+            if if_f0 == 1:
+                net_g = SynthesizerTrnMs256NSFsid(*cpt["config"], is_half=config.is_half)
+            else:
+                net_g = SynthesizerTrnMs256NSFsid_nono(*cpt["config"])
+            model_version = "V1"
+        elif version == "v2":
+            if if_f0 == 1:
+                net_g = SynthesizerTrnMs768NSFsid(*cpt["config"], is_half=config.is_half)
+            else:
+                net_g = SynthesizerTrnMs768NSFsid_nono(*cpt["config"])
+            model_version = "V2"
+        del net_g.enc_q
+        print(net_g.load_state_dict(cpt["weight"], strict=False))
+        net_g.eval().to(config.device)
+        if config.is_half:
+            net_g = net_g.half()
+        else:
+            net_g = net_g.float()
+        vc = VC(tgt_sr, config)
+        print(f"Model loaded: {character_name} / {info['feature_retrieval_library']} | ({model_version})")
+        models.append((character_name, model_title, model_author, model_cover, model_version, create_vc_fn(model_name, tgt_sr, net_g, vc, if_f0, version, model_index)))
+    return models
+
+def cut_vocal_and_inst(url, audio_provider, split_model):
+    if url != "":
+        if not os.path.exists("dl_audio"):
+            os.mkdir("dl_audio")
+        if audio_provider == "Youtube":
+            ydl_opts = {
+            'noplaylist': True,
+            'format': 'bestaudio/best',
+            'postprocessors': [{
+                'key': 'FFmpegExtractAudio',
+                'preferredcodec': 'wav',
+            }],
+            "outtmpl": 'dl_audio/youtube_audio',
+            }
+            with yt_dlp.YoutubeDL(ydl_opts) as ydl:
+                ydl.download([url])
+            audio_path = "dl_audio/youtube_audio.wav"
+        if split_model == "htdemucs":
+            command = f"demucs --two-stems=vocals {audio_path} -o output"
+            result = subprocess.run(command.split(), stdout=subprocess.PIPE)
+            print(result.stdout.decode())
+            return "output/htdemucs/youtube_audio/vocals.wav", "output/htdemucs/youtube_audio/no_vocals.wav", audio_path, "output/htdemucs/youtube_audio/vocals.wav"
+        else:
+            command = f"demucs --two-stems=vocals -n mdx_extra_q {audio_path} -o output"
+            result = subprocess.run(command.split(), stdout=subprocess.PIPE)
+            print(result.stdout.decode())
+            return "output/mdx_extra_q/youtube_audio/vocals.wav", "output/mdx_extra_q/youtube_audio/no_vocals.wav", audio_path, "output/mdx_extra_q/youtube_audio/vocals.wav"
+    else:
+        raise gr.Error("URL Required!")
+        return None, None, None, None
+
+def combine_vocal_and_inst(audio_data, audio_volume, split_model):
+    if not os.path.exists("output/result"):
+        os.mkdir("output/result")
+    vocal_path = "output/result/output.wav"
+    output_path = "output/result/combine.mp3"
+    if split_model == "htdemucs":
+        inst_path = "output/htdemucs/youtube_audio/no_vocals.wav"
+    else:
+        inst_path = "output/mdx_extra_q/youtube_audio/no_vocals.wav"
+    with wave.open(vocal_path, "w") as wave_file:
+        wave_file.setnchannels(1) 
+        wave_file.setsampwidth(2)
+        wave_file.setframerate(audio_data[0])
+        wave_file.writeframes(audio_data[1].tobytes())
+    command =  f'ffmpeg -y -i {inst_path} -i {vocal_path} -filter_complex [1:a]volume={audio_volume}dB[v];[0:a][v]amix=inputs=2:duration=longest -b:a 320k -c:a libmp3lame {output_path}'
+    result = subprocess.run(command.split(), stdout=subprocess.PIPE)
+    print(result.stdout.decode())
+    return output_path
+
+def load_hubert():
+    global hubert_model
+    models, _, _ = checkpoint_utils.load_model_ensemble_and_task(
+        ["hubert_base.pt"],
+        suffix="",
+    )
+    hubert_model = models[0]
+    hubert_model = hubert_model.to(config.device)
+    if config.is_half:
+        hubert_model = hubert_model.half()
+    else:
+        hubert_model = hubert_model.float()
+    hubert_model.eval()
+
+def change_audio_mode(vc_audio_mode):
+    if vc_audio_mode == "Input path":
+        return (
+            # Input & Upload
+            gr.Textbox.update(visible=True),
+            gr.Checkbox.update(visible=False),
+            gr.Audio.update(visible=False),
+            # Youtube
+            gr.Dropdown.update(visible=False),
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False),
+            gr.Button.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Slider.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Button.update(visible=False),
+            # TTS
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False)
+        )
+    elif vc_audio_mode == "Upload audio":
+        return (
+            # Input & Upload
+            gr.Textbox.update(visible=False),
+            gr.Checkbox.update(visible=True),
+            gr.Audio.update(visible=True),
+            # Youtube
+            gr.Dropdown.update(visible=False),
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False),
+            gr.Button.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Slider.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Button.update(visible=False),
+            # TTS
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False)
+        )
+    elif vc_audio_mode == "Youtube":
+        return (
+            # Input & Upload
+            gr.Textbox.update(visible=False),
+            gr.Checkbox.update(visible=False),
+            gr.Audio.update(visible=False),
+            # Youtube
+            gr.Dropdown.update(visible=True),
+            gr.Textbox.update(visible=True),
+            gr.Dropdown.update(visible=True),
+            gr.Button.update(visible=True),
+            gr.Audio.update(visible=True),
+            gr.Audio.update(visible=True),
+            gr.Audio.update(visible=True),
+            gr.Slider.update(visible=True),
+            gr.Audio.update(visible=True),
+            gr.Button.update(visible=True),
+            # TTS
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False)
+        )
+    elif vc_audio_mode == "TTS Audio":
+        return (
+            # Input & Upload
+            gr.Textbox.update(visible=False),
+            gr.Checkbox.update(visible=False),
+            gr.Audio.update(visible=False),
+            # Youtube
+            gr.Dropdown.update(visible=False),
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False),
+            gr.Button.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Slider.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Button.update(visible=False),
+            # TTS
+            gr.Textbox.update(visible=True),
+            gr.Dropdown.update(visible=True)
+        )
+    else:
+        return (
+            # Input & Upload
+            gr.Textbox.update(visible=False),
+            gr.Checkbox.update(visible=True),
+            gr.Audio.update(visible=True),
+            # Youtube
+            gr.Dropdown.update(visible=False),
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False),
+            gr.Button.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Slider.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Button.update(visible=False),
+            # TTS
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False)
+        )
+
+def use_microphone(microphone):
+    if microphone == True:
+        return gr.Audio.update(source="microphone")
+    else:
+        return gr.Audio.update(source="upload")
+
+if __name__ == '__main__':
+    load_hubert()
+    models = load_model()
+    tts_voice_list = asyncio.new_event_loop().run_until_complete(edge_tts.list_voices())
+    voices = [f"{v['ShortName']}-{v['Gender']}" for v in tts_voice_list]
+    with gr.Blocks() as app:
+        gr.Markdown(
+            "# <center> Combined Genshin Impact RVC Models\n"
+            "## <center> The input audio should be clean and pure voice without background music.\n"
+            "### <center> It is recommended to use google colab for more features. \n"
+            "[![image](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1Tgr6q9kKiB5P37rUitrB3CsNl8JP9iQZ?usp=sharing)\n\n"
+            "[![Original Repo](https://badgen.net/badge/icon/github?icon=github&label=Original%20Repo)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI)"
+        )
+        with gr.Tabs():
+            for (name, title, author, cover, model_version, vc_fn) in models:
+                with gr.TabItem(name):
+                    with gr.Row():
+                        gr.Markdown(
+                            '<div align="center">'
+                            f'<div>{title}</div>\n'+
+                            f'<div>RVC {model_version} Model</div>\n'+
+                            (f'<div>Model author: {author}</div>' if author else "")+
+                            (f'<img style="width:auto;height:300px;" src="file/{cover}">' if cover else "")+
+                            '</div>'
+                        )
+                    with gr.Row():
+                        with gr.Column():
+                            vc_audio_mode = gr.Dropdown(label="Input voice", choices=audio_mode, allow_custom_value=False, value="Upload audio")
+                            # Input
+                            vc_input = gr.Textbox(label="Input audio path", visible=False)
+                            # Upload
+                            vc_microphone_mode = gr.Checkbox(label="Use Microphone", value=False, visible=True, interactive=True)
+                            vc_upload = gr.Audio(label="Upload audio file", source="upload", visible=True, interactive=True)
+                            # Youtube
+                            vc_download_audio = gr.Dropdown(label="Provider", choices=["Youtube"], allow_custom_value=False, visible=False, value="Youtube", info="Select provider (Default: Youtube)")
+                            vc_link = gr.Textbox(label="Youtube URL", visible=False, info="Example: https://www.youtube.com/watch?v=Nc0sB1Bmf-A", placeholder="https://www.youtube.com/watch?v=...")
+                            vc_split_model = gr.Dropdown(label="Splitter Model", choices=["htdemucs", "mdx_extra_q"], allow_custom_value=False, visible=False, value="htdemucs", info="Select the splitter model (Default: htdemucs)")
+                            vc_split = gr.Button("Split Audio", variant="primary", visible=False)
+                            vc_vocal_preview = gr.Audio(label="Vocal Preview", visible=False)
+                            vc_inst_preview = gr.Audio(label="Instrumental Preview", visible=False)
+                            vc_audio_preview = gr.Audio(label="Audio Preview", visible=False)
+                            # TTS
+                            tts_text = gr.Textbox(visible=False, label="TTS text", info="Text to speech input")
+                            tts_voice = gr.Dropdown(label="Edge-tts speaker", choices=voices, visible=False, allow_custom_value=False, value="en-US-AnaNeural-Female")
+                        with gr.Column():
+                            vc_transform0 = gr.Number(label="Transpose", value=0, info='Type "12" to change from male to female voice. Type "-12" to change female to male voice')
+                            f0method0 = gr.Radio(
+                                label="Pitch extraction algorithm",
+                                info=f0method_info,
+                                choices=f0method_mode,
+                                value="pm",
+                                interactive=True
+                            )
+                            index_rate1 = gr.Slider(
+                                minimum=0,
+                                maximum=1,
+                                label="Retrieval feature ratio",
+                                info="(Default: 0.7)",
+                                value=0.7,
+                                interactive=True,
+                            )
+                            filter_radius0 = gr.Slider(
+                                minimum=0,
+                                maximum=7,
+                                label="Apply Median Filtering",
+                                info="The value represents the filter radius and can reduce breathiness.",
+                                value=3,
+                                step=1,
+                                interactive=True,
+                            )
+                            resample_sr0 = gr.Slider(
+                                minimum=0,
+                                maximum=48000,
+                                label="Resample the output audio",
+                                info="Resample the output audio in post-processing to the final sample rate. Set to 0 for no resampling",
+                                value=0,
+                                step=1,
+                                interactive=True,
+                            )
+                            rms_mix_rate0 = gr.Slider(
+                                minimum=0,
+                                maximum=1,
+                                label="Volume Envelope",
+                                info="Use the volume envelope of the input to replace or mix with the volume envelope of the output. The closer the ratio is to 1, the more the output envelope is used",
+                                value=1,
+                                interactive=True,
+                            )
+                            protect0 = gr.Slider(
+                                minimum=0,
+                                maximum=0.5,
+                                label="Voice Protection",
+                                info="Protect voiceless consonants and breath sounds to prevent artifacts such as tearing in electronic music. Set to 0.5 to disable. Decrease the value to increase protection, but it may reduce indexing accuracy",
+                                value=0.5,
+                                step=0.01,
+                                interactive=True,
+                            )
+                        with gr.Column():
+                            vc_log = gr.Textbox(label="Output Information", interactive=False)
+                            vc_output = gr.Audio(label="Output Audio", interactive=False)
+                            vc_convert = gr.Button("Convert", variant="primary")
+                            vc_volume = gr.Slider(
+                                minimum=0,
+                                maximum=10,
+                                label="Vocal volume",
+                                value=4,
+                                interactive=True,
+                                step=1,
+                                info="Adjust vocal volume (Default: 4}",
+                                visible=False
+                            )
+                            vc_combined_output = gr.Audio(label="Output Combined Audio", visible=False)
+                            vc_combine =  gr.Button("Combine",variant="primary", visible=False)
+                vc_convert.click(
+                    fn=vc_fn, 
+                    inputs=[
+                        vc_audio_mode,
+                        vc_input,
+                        vc_upload,
+                        tts_text,
+                        tts_voice,
+                        vc_transform0,
+                        f0method0,
+                        index_rate1,
+                        filter_radius0,
+                        resample_sr0,
+                        rms_mix_rate0,
+                        protect0,
+                    ], 
+                    outputs=[vc_log ,vc_output]
+                )
+                vc_split.click(
+                    fn=cut_vocal_and_inst, 
+                    inputs=[vc_link, vc_download_audio, vc_split_model], 
+                    outputs=[vc_vocal_preview, vc_inst_preview, vc_audio_preview, vc_input]
+                )
+                vc_combine.click(
+                    fn=combine_vocal_and_inst,
+                    inputs=[vc_output, vc_volume, vc_split_model],
+                    outputs=[vc_combined_output]
+                )
+                vc_microphone_mode.change(
+                    fn=use_microphone,
+                    inputs=vc_microphone_mode,
+                    outputs=vc_upload
+                )
+                vc_audio_mode.change(
+                    fn=change_audio_mode,
+                    inputs=[vc_audio_mode],
+                    outputs=[
+                        vc_input,
+                        vc_microphone_mode,
+                        vc_upload,
+                        vc_download_audio,
+                        vc_link,
+                        vc_split_model,
+                        vc_split,
+                        vc_vocal_preview,
+                        vc_inst_preview,
+                        vc_audio_preview,
+                        vc_volume,
+                        vc_combined_output,
+                        vc_combine,
+                        tts_text,
+                        tts_voice
+                    ]
+                )
+        app.queue(concurrency_count=1, max_size=20, api_open=config.api).launch(share=config.colab)

app.py

@@ -0,0 +1,504 @@
+import os
+import glob
+import json
+import argparse
+import traceback
+import logging
+import gradio as gr
+import numpy as np
+import librosa
+import torch
+import asyncio
+import edge_tts
+import yt_dlp
+import ffmpeg
+import subprocess
+import sys
+import io
+import wave
+from datetime import datetime
+from fairseq import checkpoint_utils
+from lib.infer_pack.models import (
+    SynthesizerTrnMs256NSFsid,
+    SynthesizerTrnMs256NSFsid_nono,
+    SynthesizerTrnMs768NSFsid,
+    SynthesizerTrnMs768NSFsid_nono,
+)
+from vc_infer_pipeline import VC
+from config import Config
+config = Config()
+logging.getLogger("numba").setLevel(logging.WARNING)
+limitation = os.getenv("SYSTEM") == "spaces"  # limit audio length in huggingface spaces
+
+audio_mode = []
+f0method_mode = []
+f0method_info = ""
+
+if limitation is True:
+    audio_mode = ["Upload audio", "TTS Audio"]
+    f0method_mode = ["pm", "harvest"]
+    f0method_info = "PM is fast, Harvest is good but extremely slow, Rvmpe is alternative to harvest (might be better). (Default: PM)"
+else:
+    audio_mode = ["Input path", "Upload audio", "Youtube", "TTS Audio"]
+    f0method_mode = ["pm", "harvest", "crepe"]
+    f0method_info = "PM is fast, Harvest is good but extremely slow, Rvmpe is alternative to harvest (might be better), and Crepe effect is good but requires GPU (Default: PM)"
+
+if os.path.isfile("rmvpe.pt"):
+    f0method_mode.insert(2, "rmvpe")
+
+def create_vc_fn(model_name, tgt_sr, net_g, vc, if_f0, version, file_index):
+    def vc_fn(
+        vc_audio_mode,
+        vc_input, 
+        vc_upload,
+        tts_text,
+        tts_voice,
+        f0_up_key,
+        f0_method,
+        index_rate,
+        filter_radius,
+        resample_sr,
+        rms_mix_rate,
+        protect
+    ):
+        try:
+            print(f"Converting using {model_name}...")
+            if vc_audio_mode == "Input path" or "Youtube" and vc_input != "":
+                audio, sr = librosa.load(vc_input, sr=16000, mono=True)
+            elif vc_audio_mode == "Upload audio":
+                if vc_upload is None:
+                    return "You need to upload an audio", None
+                sampling_rate, audio = vc_upload
+                duration = audio.shape[0] / sampling_rate
+                if duration > 20 and limitation:
+                    return "Please upload an audio file that is less than 20 seconds. If you need to generate a longer audio file, please use Colab.", None
+                audio = (audio / np.iinfo(audio.dtype).max).astype(np.float32)
+                if len(audio.shape) > 1:
+                    audio = librosa.to_mono(audio.transpose(1, 0))
+                if sampling_rate != 16000:
+                    audio = librosa.resample(audio, orig_sr=sampling_rate, target_sr=16000)
+            elif vc_audio_mode == "TTS Audio":
+                if len(tts_text) > 100 and limitation:
+                    return "Text is too long", None
+                if tts_text is None or tts_voice is None:
+                    return "You need to enter text and select a voice", None
+                asyncio.run(edge_tts.Communicate(tts_text, "-".join(tts_voice.split('-')[:-1])).save("tts.mp3"))
+                audio, sr = librosa.load("tts.mp3", sr=16000, mono=True)
+                vc_input = "tts.mp3"
+            times = [0, 0, 0]
+            f0_up_key = int(f0_up_key)
+            audio_opt = vc.pipeline(
+                hubert_model,
+                net_g,
+                0,
+                audio,
+                vc_input,
+                times,
+                f0_up_key,
+                f0_method,
+                file_index,
+                # file_big_npy,
+                index_rate,
+                if_f0,
+                filter_radius,
+                tgt_sr,
+                resample_sr,
+                rms_mix_rate,
+                version,
+                protect,
+                f0_file=None,
+            )
+            info = f"[{datetime.now().strftime('%Y-%m-%d %H:%M')}]: npy: {times[0]}, f0: {times[1]}s, infer: {times[2]}s"
+            print(info)
+            return info, (tgt_sr, audio_opt)
+        except:
+            info = traceback.format_exc()
+            print(f"{model_name} | {info}")
+            return info, (tgt_sr, audio_opt)
+    return vc_fn
+
+def load_model():
+    models = []
+    with open(f"weights/model_info.json", "r", encoding="utf-8") as f:
+        models_info = json.load(f)
+    for character_name, info in models_info.items():
+        if not info['enable']:
+            continue
+        model_title = info['title']
+        model_name = info['model_path']
+        model_author = info.get("author", None)
+        model_cover = f"weights/{character_name}/{info['cover']}"
+        model_index = f"weights/{character_name}/{info['feature_retrieval_library']}"
+        cpt = torch.load(f"weights/{character_name}/{model_name}", map_location="cpu")
+        tgt_sr = cpt["config"][-1]
+        cpt["config"][-3] = cpt["weight"]["emb_g.weight"].shape[0]  # n_spk
+        if_f0 = cpt.get("f0", 1)
+        version = cpt.get("version", "v1")
+        if version == "v1":
+            if if_f0 == 1:
+                net_g = SynthesizerTrnMs256NSFsid(*cpt["config"], is_half=config.is_half)
+            else:
+                net_g = SynthesizerTrnMs256NSFsid_nono(*cpt["config"])
+            model_version = "V1"
+        elif version == "v2":
+            if if_f0 == 1:
+                net_g = SynthesizerTrnMs768NSFsid(*cpt["config"], is_half=config.is_half)
+            else:
+                net_g = SynthesizerTrnMs768NSFsid_nono(*cpt["config"])
+            model_version = "V2"
+        del net_g.enc_q
+        print(net_g.load_state_dict(cpt["weight"], strict=False))
+        net_g.eval().to(config.device)
+        if config.is_half:
+            net_g = net_g.half()
+        else:
+            net_g = net_g.float()
+        vc = VC(tgt_sr, config)
+        print(f"Model loaded: {character_name} / {info['feature_retrieval_library']} | ({model_version})")
+        models.append((character_name, model_title, model_author, model_cover, model_version, create_vc_fn(model_title, tgt_sr, net_g, vc, if_f0, version, model_index)))
+    return models
+
+def cut_vocal_and_inst(url, audio_provider, split_model):
+    if url != "":
+        if not os.path.exists("dl_audio"):
+            os.mkdir("dl_audio")
+        if audio_provider == "Youtube":
+            ydl_opts = {
+            'noplaylist': True,
+            'format': 'bestaudio/best',
+            'postprocessors': [{
+                'key': 'FFmpegExtractAudio',
+                'preferredcodec': 'wav',
+            }],
+            "outtmpl": 'dl_audio/youtube_audio',
+            }
+            with yt_dlp.YoutubeDL(ydl_opts) as ydl:
+                ydl.download([url])
+            audio_path = "dl_audio/youtube_audio.wav"
+        if split_model == "htdemucs":
+            command = f"demucs --two-stems=vocals {audio_path} -o output"
+            result = subprocess.run(command.split(), stdout=subprocess.PIPE)
+            print(result.stdout.decode())
+            return "output/htdemucs/youtube_audio/vocals.wav", "output/htdemucs/youtube_audio/no_vocals.wav", audio_path, "output/htdemucs/youtube_audio/vocals.wav"
+        else:
+            command = f"demucs --two-stems=vocals -n mdx_extra_q {audio_path} -o output"
+            result = subprocess.run(command.split(), stdout=subprocess.PIPE)
+            print(result.stdout.decode())
+            return "output/mdx_extra_q/youtube_audio/vocals.wav", "output/mdx_extra_q/youtube_audio/no_vocals.wav", audio_path, "output/mdx_extra_q/youtube_audio/vocals.wav"
+    else:
+        raise gr.Error("URL Required!")
+        return None, None, None, None
+
+def combine_vocal_and_inst(audio_data, audio_volume, split_model):
+    if not os.path.exists("output/result"):
+        os.mkdir("output/result")
+    vocal_path = "output/result/output.wav"
+    output_path = "output/result/combine.mp3"
+    if split_model == "htdemucs":
+        inst_path = "output/htdemucs/youtube_audio/no_vocals.wav"
+    else:
+        inst_path = "output/mdx_extra_q/youtube_audio/no_vocals.wav"
+    with wave.open(vocal_path, "w") as wave_file:
+        wave_file.setnchannels(1) 
+        wave_file.setsampwidth(2)
+        wave_file.setframerate(audio_data[0])
+        wave_file.writeframes(audio_data[1].tobytes())
+    command =  f'ffmpeg -y -i {inst_path} -i {vocal_path} -filter_complex [1:a]volume={audio_volume}dB[v];[0:a][v]amix=inputs=2:duration=longest -b:a 320k -c:a libmp3lame {output_path}'
+    result = subprocess.run(command.split(), stdout=subprocess.PIPE)
+    print(result.stdout.decode())
+    return output_path
+
+def load_hubert():
+    global hubert_model
+    models, _, _ = checkpoint_utils.load_model_ensemble_and_task(
+        ["hubert_base.pt"],
+        suffix="",
+    )
+    hubert_model = models[0]
+    hubert_model = hubert_model.to(config.device)
+    if config.is_half:
+        hubert_model = hubert_model.half()
+    else:
+        hubert_model = hubert_model.float()
+    hubert_model.eval()
+
+def change_audio_mode(vc_audio_mode):
+    if vc_audio_mode == "Input path":
+        return (
+            # Input & Upload
+            gr.Textbox.update(visible=True),
+            gr.Checkbox.update(visible=False),
+            gr.Audio.update(visible=False),
+            # Youtube
+            gr.Dropdown.update(visible=False),
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False),
+            gr.Button.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Slider.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Button.update(visible=False),
+            # TTS
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False)
+        )
+    elif vc_audio_mode == "Upload audio":
+        return (
+            # Input & Upload
+            gr.Textbox.update(visible=False),
+            gr.Checkbox.update(visible=True),
+            gr.Audio.update(visible=True),
+            # Youtube
+            gr.Dropdown.update(visible=False),
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False),
+            gr.Button.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Slider.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Button.update(visible=False),
+            # TTS
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False)
+        )
+    elif vc_audio_mode == "Youtube":
+        return (
+            # Input & Upload
+            gr.Textbox.update(visible=False),
+            gr.Checkbox.update(visible=False),
+            gr.Audio.update(visible=False),
+            # Youtube
+            gr.Dropdown.update(visible=True),
+            gr.Textbox.update(visible=True),
+            gr.Dropdown.update(visible=True),
+            gr.Button.update(visible=True),
+            gr.Audio.update(visible=True),
+            gr.Audio.update(visible=True),
+            gr.Audio.update(visible=True),
+            gr.Slider.update(visible=True),
+            gr.Audio.update(visible=True),
+            gr.Button.update(visible=True),
+            # TTS
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False)
+        )
+    elif vc_audio_mode == "TTS Audio":
+        return (
+            # Input & Upload
+            gr.Textbox.update(visible=False),
+            gr.Checkbox.update(visible=False),
+            gr.Audio.update(visible=False),
+            # Youtube
+            gr.Dropdown.update(visible=False),
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False),
+            gr.Button.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Slider.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Button.update(visible=False),
+            # TTS
+            gr.Textbox.update(visible=True),
+            gr.Dropdown.update(visible=True)
+        )
+    else:
+        return (
+            # Input & Upload
+            gr.Textbox.update(visible=False),
+            gr.Checkbox.update(visible=True),
+            gr.Audio.update(visible=True),
+            # Youtube
+            gr.Dropdown.update(visible=False),
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False),
+            gr.Button.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Slider.update(visible=False),
+            gr.Audio.update(visible=False),
+            gr.Button.update(visible=False),
+            # TTS
+            gr.Textbox.update(visible=False),
+            gr.Dropdown.update(visible=False)
+        )
+
+def use_microphone(microphone):
+    if microphone == True:
+        return gr.Audio.update(source="microphone")
+    else:
+        return gr.Audio.update(source="upload")
+
+if __name__ == '__main__':
+    load_hubert()
+    models = load_model()
+    tts_voice_list = asyncio.new_event_loop().run_until_complete(edge_tts.list_voices())
+    voices = [f"{v['ShortName']}-{v['Gender']}" for v in tts_voice_list]
+    with gr.Blocks() as app:
+        gr.Markdown(
+            "# <center> Combined Genshin Impact RVC Models\n"
+            "## <center> The input audio should be clean and pure voice without background music.\n"
+            "### <center> It is recommended to use google colab for more features. \n"
+            "[![image](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1Tgr6q9kKiB5P37rUitrB3CsNl8JP9iQZ?usp=sharing)\n\n"
+            "[![Original Repo](https://badgen.net/badge/icon/github?icon=github&label=Original%20Repo)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI)"
+        )
+        with gr.Tabs():
+            for (name, title, author, cover, model_version, vc_fn) in models:
+                with gr.TabItem(name):
+                    with gr.Row():
+                        gr.Markdown(
+                            '<div align="center">'
+                            f'<div>{title}</div>\n'+
+                            f'<div>RVC {model_version} Model</div>\n'+
+                            (f'<div>Model author: {author}</div>' if author else "")+
+                            (f'<img style="width:auto;height:300px;" src="file/{cover}">' if cover else "")+
+                            '</div>'
+                        )
+                    with gr.Row():
+                        with gr.Column():
+                            vc_audio_mode = gr.Dropdown(label="Input voice", choices=audio_mode, allow_custom_value=False, value="Upload audio")
+                            # Input
+                            vc_input = gr.Textbox(label="Input audio path", visible=False)
+                            # Upload
+                            vc_microphone_mode = gr.Checkbox(label="Use Microphone", value=False, visible=True, interactive=True)
+                            vc_upload = gr.Audio(label="Upload audio file", source="upload", visible=True, interactive=True)
+                            # Youtube
+                            vc_download_audio = gr.Dropdown(label="Provider", choices=["Youtube"], allow_custom_value=False, visible=False, value="Youtube", info="Select provider (Default: Youtube)")
+                            vc_link = gr.Textbox(label="Youtube URL", visible=False, info="Example: https://www.youtube.com/watch?v=Nc0sB1Bmf-A", placeholder="https://www.youtube.com/watch?v=...")
+                            vc_split_model = gr.Dropdown(label="Splitter Model", choices=["htdemucs", "mdx_extra_q"], allow_custom_value=False, visible=False, value="htdemucs", info="Select the splitter model (Default: htdemucs)")
+                            vc_split = gr.Button("Split Audio", variant="primary", visible=False)
+                            vc_vocal_preview = gr.Audio(label="Vocal Preview", visible=False)
+                            vc_inst_preview = gr.Audio(label="Instrumental Preview", visible=False)
+                            vc_audio_preview = gr.Audio(label="Audio Preview", visible=False)
+                            # TTS
+                            tts_text = gr.Textbox(visible=False, label="TTS text", info="Text to speech input")
+                            tts_voice = gr.Dropdown(label="Edge-tts speaker", choices=voices, visible=False, allow_custom_value=False, value="en-US-AnaNeural-Female")
+                        with gr.Column():
+                            vc_transform0 = gr.Number(label="Transpose", value=0, info='Type "12" to change from male to female voice. Type "-12" to change female to male voice')
+                            f0method0 = gr.Radio(
+                                label="Pitch extraction algorithm",
+                                info=f0method_info,
+                                choices=f0method_mode,
+                                value="pm",
+                                interactive=True
+                            )
+                            index_rate1 = gr.Slider(
+                                minimum=0,
+                                maximum=1,
+                                label="Retrieval feature ratio",
+                                info="(Default: 0.7)",
+                                value=0.7,
+                                interactive=True,
+                            )
+                            filter_radius0 = gr.Slider(
+                                minimum=0,
+                                maximum=7,
+                                label="Apply Median Filtering",
+                                info="The value represents the filter radius and can reduce breathiness.",
+                                value=3,
+                                step=1,
+                                interactive=True,
+                            )
+                            resample_sr0 = gr.Slider(
+                                minimum=0,
+                                maximum=48000,
+                                label="Resample the output audio",
+                                info="Resample the output audio in post-processing to the final sample rate. Set to 0 for no resampling",
+                                value=0,
+                                step=1,
+                                interactive=True,
+                            )
+                            rms_mix_rate0 = gr.Slider(
+                                minimum=0,
+                                maximum=1,
+                                label="Volume Envelope",
+                                info="Use the volume envelope of the input to replace or mix with the volume envelope of the output. The closer the ratio is to 1, the more the output envelope is used",
+                                value=1,
+                                interactive=True,
+                            )
+                            protect0 = gr.Slider(
+                                minimum=0,
+                                maximum=0.5,
+                                label="Voice Protection",
+                                info="Protect voiceless consonants and breath sounds to prevent artifacts such as tearing in electronic music. Set to 0.5 to disable. Decrease the value to increase protection, but it may reduce indexing accuracy",
+                                value=0.5,
+                                step=0.01,
+                                interactive=True,
+                            )
+                        with gr.Column():
+                            vc_log = gr.Textbox(label="Output Information", interactive=False)
+                            vc_output = gr.Audio(label="Output Audio", interactive=False)
+                            vc_convert = gr.Button("Convert", variant="primary")
+                            vc_volume = gr.Slider(
+                                minimum=0,
+                                maximum=10,
+                                label="Vocal volume",
+                                value=4,
+                                interactive=True,
+                                step=1,
+                                info="Adjust vocal volume (Default: 4}",
+                                visible=False
+                            )
+                            vc_combined_output = gr.Audio(label="Output Combined Audio", visible=False)
+                            vc_combine =  gr.Button("Combine",variant="primary", visible=False)
+                vc_convert.click(
+                    fn=vc_fn, 
+                    inputs=[
+                        vc_audio_mode,
+                        vc_input,
+                        vc_upload,
+                        tts_text,
+                        tts_voice,
+                        vc_transform0,
+                        f0method0,
+                        index_rate1,
+                        filter_radius0,
+                        resample_sr0,
+                        rms_mix_rate0,
+                        protect0,
+                    ], 
+                    outputs=[vc_log ,vc_output]
+                )
+                vc_split.click(
+                    fn=cut_vocal_and_inst, 
+                    inputs=[vc_link, vc_download_audio, vc_split_model], 
+                    outputs=[vc_vocal_preview, vc_inst_preview, vc_audio_preview, vc_input]
+                )
+                vc_combine.click(
+                    fn=combine_vocal_and_inst,
+                    inputs=[vc_output, vc_volume, vc_split_model],
+                    outputs=[vc_combined_output]
+                )
+                vc_microphone_mode.change(
+                    fn=use_microphone,
+                    inputs=vc_microphone_mode,
+                    outputs=vc_upload
+                )
+                vc_audio_mode.change(
+                    fn=change_audio_mode,
+                    inputs=[vc_audio_mode],
+                    outputs=[
+                        vc_input,
+                        vc_microphone_mode, 
+                        vc_upload,
+                        vc_download_audio,
+                        vc_link,
+                        vc_split_model,
+                        vc_split,
+                        vc_vocal_preview,
+                        vc_inst_preview,
+                        vc_audio_preview,
+                        vc_volume,
+                        vc_combined_output,
+                        vc_combine,
+                        tts_text,
+                        tts_voice
+                    ]
+                )
+        app.queue(concurrency_count=1, max_size=20, api_open=config.api).launch(share=config.colab)

config.py

@@ -0,0 +1,117 @@
+import argparse
+import sys
+import torch
+from multiprocessing import cpu_count
+
+class Config:
+    def __init__(self):
+        self.device = "cuda:0"
+        self.is_half = True
+        self.n_cpu = 0
+        self.gpu_name = None
+        self.gpu_mem = None
+        (
+            self.python_cmd,
+            self.listen_port,
+            self.colab,
+            self.noparallel,
+            self.noautoopen,
+            self.api
+        ) = self.arg_parse()
+        self.x_pad, self.x_query, self.x_center, self.x_max = self.device_config()
+
+    @staticmethod
+    def arg_parse() -> tuple:
+        exe = sys.executable or "python"
+        parser = argparse.ArgumentParser()
+        parser.add_argument("--port", type=int, default=7865, help="Listen port")
+        parser.add_argument("--pycmd", type=str, default=exe, help="Python command")
+        parser.add_argument("--colab", action="store_true", help="Launch in colab")
+        parser.add_argument(
+            "--noparallel", action="store_true", help="Disable parallel processing"
+        )
+        parser.add_argument(
+            "--noautoopen",
+            action="store_true",
+            help="Do not open in browser automatically",
+        )
+        parser.add_argument("--api", action="store_true", help="Launch with api")
+        cmd_opts = parser.parse_args()
+
+        cmd_opts.port = cmd_opts.port if 0 <= cmd_opts.port <= 65535 else 7865
+
+        return (
+            cmd_opts.pycmd,
+            cmd_opts.port,
+            cmd_opts.colab,
+            cmd_opts.noparallel,
+            cmd_opts.noautoopen,
+            cmd_opts.api
+        )
+
+    # has_mps is only available in nightly pytorch (for now) and MasOS 12.3+.
+    # check `getattr` and try it for compatibility
+    @staticmethod
+    def has_mps() -> bool:
+        if not torch.backends.mps.is_available():
+            return False
+        try:
+            torch.zeros(1).to(torch.device("mps"))
+            return True
+        except Exception:
+            return False
+
+    def device_config(self) -> tuple:
+        if torch.cuda.is_available():
+            i_device = int(self.device.split(":")[-1])
+            self.gpu_name = torch.cuda.get_device_name(i_device)
+            if (
+                ("16" in self.gpu_name and "V100" not in self.gpu_name.upper())
+                or "P40" in self.gpu_name.upper()
+                or "1060" in self.gpu_name
+                or "1070" in self.gpu_name
+                or "1080" in self.gpu_name
+            ):
+                print("Found GPU", self.gpu_name, ", force to fp32")
+                self.is_half = False
+            else:
+                print("Found GPU", self.gpu_name)
+            self.gpu_mem = int(
+                torch.cuda.get_device_properties(i_device).total_memory
+                / 1024
+                / 1024
+                / 1024
+                + 0.4
+            )
+        elif self.has_mps():
+            print("No supported Nvidia GPU found, use MPS instead")
+            self.device = "mps"
+            self.is_half = False
+        else:
+            print("No supported Nvidia GPU found, use CPU instead")
+            self.device = "cpu"
+            self.is_half = False
+
+        if self.n_cpu == 0:
+            self.n_cpu = cpu_count()
+
+        if self.is_half:
+            # 6G显存配置
+            x_pad = 3
+            x_query = 10
+            x_center = 60
+            x_max = 65
+        else:
+            # 5G显存配置
+            x_pad = 1
+            x_query = 6
+            x_center = 38
+            x_max = 41
+
+        if self.gpu_mem != None and self.gpu_mem <= 4:
+            x_pad = 1
+            x_query = 5
+            x_center = 30
+            x_max = 32
+
+        return x_pad, x_query, x_center, x_max

gitattributes.txt

@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.index filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text

gitignore.txt

@@ -0,0 +1,382 @@
+## Ignore Visual Studio temporary files, build results, and
+## files generated by popular Visual Studio add-ons.
+##
+## Get latest from https://github.com/github/gitignore/blob/master/VisualStudio.gitignore
+
+# User-specific files
+*.rsuser
+*.suo
+*.user
+*.userosscache
+*.sln.docstates
+
+# User-specific files (MonoDevelop/Xamarin Studio)
+*.userprefs
+
+# Mono auto generated files
+mono_crash.*
+
+# Build results
+[Dd]ebug/
+[Dd]ebugPublic/
+[Rr]elease/
+[Rr]eleases/
+x64/
+x86/
+[Ww][Ii][Nn]32/
+[Aa][Rr][Mm]/
+[Aa][Rr][Mm]64/
+bld/
+[Bb]in/
+[Oo]bj/
+[Oo]ut/
+[Ll]og/
+[Ll]ogs/
+infer_pack\__pycache__
+# Visual Studio 2015/2017 cache/options directory
+.vs/
+# Uncomment if you have tasks that create the project's static files in wwwroot
+#wwwroot/
+
+# Visual Studio 2017 auto generated files
+Generated\ Files/
+
+# MSTest test Results
+[Tt]est[Rr]esult*/
+[Bb]uild[Ll]og.*
+
+# NUnit
+*.VisualState.xml
+TestResult.xml
+nunit-*.xml
+
+# Build Results of an ATL Project
+[Dd]ebugPS/
+[Rr]eleasePS/
+dlldata.c
+
+# Benchmark Results
+BenchmarkDotNet.Artifacts/
+
+# .NET Core
+project.lock.json
+project.fragment.lock.json
+artifacts/
+
+# ASP.NET Scaffolding
+ScaffoldingReadMe.txt
+
+# StyleCop
+StyleCopReport.xml
+
+# Files built by Visual Studio
+*_i.c
+*_p.c
+*_h.h
+*.ilk
+*.meta
+*.obj
+*.iobj
+*.pch
+*.pdb
+*.ipdb
+*.pgc
+*.pgd
+*.rsp
+*.sbr
+*.tlb
+*.tli
+*.tlh
+*.tmp
+*.tmp_proj
+*_wpftmp.csproj
+*.log
+*.vspscc
+*.vssscc
+.builds
+*.pidb
+*.svclog
+*.scc
+
+# Chutzpah Test files
+_Chutzpah*
+
+# Visual C++ cache files
+ipch/
+*.aps
+*.ncb
+*.opendb
+*.opensdf
+*.sdf
+*.cachefile
+*.VC.db
+*.VC.VC.opendb
+
+# Visual Studio profiler
+*.psess
+*.vsp
+*.vspx
+*.sap
+
+# Visual Studio Trace Files
+*.e2e
+
+# TFS 2012 Local Workspace
+$tf/
+
+# Guidance Automation Toolkit
+*.gpState
+
+# ReSharper is a .NET coding add-in
+_ReSharper*/
+*.[Rr]e[Ss]harper
+*.DotSettings.user
+
+# TeamCity is a build add-in
+_TeamCity*
+
+# DotCover is a Code Coverage Tool
+*.dotCover
+
+# AxoCover is a Code Coverage Tool
+.axoCover/*
+!.axoCover/settings.json
+
+# Coverlet is a free, cross platform Code Coverage Tool
+coverage*.json
+coverage*.xml
+coverage*.info
+
+# Visual Studio code coverage results
+*.coverage
+*.coveragexml
+
+# NCrunch
+_NCrunch_*
+.*crunch*.local.xml
+nCrunchTemp_*
+
+# MightyMoose
+*.mm.*
+AutoTest.Net/
+
+# Web workbench (sass)
+.sass-cache/
+
+# Installshield output folder
+[Ee]xpress/
+
+# DocProject is a documentation generator add-in
+DocProject/buildhelp/
+DocProject/Help/*.HxT
+DocProject/Help/*.HxC
+DocProject/Help/*.hhc
+DocProject/Help/*.hhk
+DocProject/Help/*.hhp
+DocProject/Help/Html2
+DocProject/Help/html
+
+# Click-Once directory
+publish/
+
+# Publish Web Output
+*.[Pp]ublish.xml
+*.azurePubxml
+# Note: Comment the next line if you want to checkin your web deploy settings,
+# but database connection strings (with potential passwords) will be unencrypted
+*.pubxml
+*.publishproj
+
+# Microsoft Azure Web App publish settings. Comment the next line if you want to
+# checkin your Azure Web App publish settings, but sensitive information contained
+# in these scripts will be unencrypted
+PublishScripts/
+
+# NuGet Packages
+*.nupkg
+# NuGet Symbol Packages
+*.snupkg
+# The packages folder can be ignored because of Package Restore
+**/[Pp]ackages/*
+# except build/, which is used as an MSBuild target.
+!**/[Pp]ackages/build/
+# Uncomment if necessary however generally it will be regenerated when needed
+#!**/[Pp]ackages/repositories.config
+# NuGet v3's project.json files produces more ignorable files
+*.nuget.props
+*.nuget.targets
+
+# Microsoft Azure Build Output
+csx/
+*.build.csdef
+
+# Microsoft Azure Emulator
+ecf/
+rcf/
+
+# Windows Store app package directories and files
+AppPackages/
+BundleArtifacts/
+Package.StoreAssociation.xml
+_pkginfo.txt
+*.appx
+*.appxbundle
+*.appxupload
+
+# Visual Studio cache files
+# files ending in .cache can be ignored
+*.[Cc]ache
+# but keep track of directories ending in .cache
+!?*.[Cc]ache/
+
+# Others
+ClientBin/
+~$*
+*~
+*.dbmdl
+*.dbproj.schemaview
+*.jfm
+*.pfx
+*.publishsettings
+orleans.codegen.cs
+
+# Including strong name files can present a security risk
+# (https://github.com/github/gitignore/pull/2483#issue-259490424)
+#*.snk
+
+# Since there are multiple workflows, uncomment next line to ignore bower_components
+# (https://github.com/github/gitignore/pull/1529#issuecomment-104372622)
+#bower_components/
+
+# RIA/Silverlight projects
+Generated_Code/
+
+# Backup & report files from converting an old project file
+# to a newer Visual Studio version. Backup files are not needed,
+# because we have git ;-)
+_UpgradeReport_Files/
+Backup*/
+UpgradeLog*.XML
+UpgradeLog*.htm
+ServiceFabricBackup/
+*.rptproj.bak
+
+# SQL Server files
+*.mdf
+*.ldf
+*.ndf
+
+# Business Intelligence projects
+*.rdl.data
+*.bim.layout
+*.bim_*.settings
+*.rptproj.rsuser
+*- [Bb]ackup.rdl
+*- [Bb]ackup ([0-9]).rdl
+*- [Bb]ackup ([0-9][0-9]).rdl
+
+# Microsoft Fakes
+FakesAssemblies/
+
+# GhostDoc plugin setting file
+*.GhostDoc.xml
+
+# Node.js Tools for Visual Studio
+.ntvs_analysis.dat
+node_modules/
+
+# Visual Studio 6 build log
+*.plg
+
+# Visual Studio 6 workspace options file
+*.opt
+
+# Visual Studio 6 auto-generated workspace file (contains which files were open etc.)
+*.vbw
+
+# Visual Studio LightSwitch build output
+**/*.HTMLClient/GeneratedArtifacts
+**/*.DesktopClient/GeneratedArtifacts
+**/*.DesktopClient/ModelManifest.xml
+**/*.Server/GeneratedArtifacts
+**/*.Server/ModelManifest.xml
+_Pvt_Extensions
+
+# Paket dependency manager
+.paket/paket.exe
+paket-files/
+
+# FAKE - F# Make
+.fake/
+
+# CodeRush personal settings
+.cr/personal
+
+# Python Tools for Visual Studio (PTVS)
+__pycache__/
+
+
+# Cake - Uncomment if you are using it
+# tools/**
+# !tools/packages.config
+
+# Tabs Studio
+*.tss
+
+# Telerik's JustMock configuration file
+*.jmconfig
+
+# BizTalk build output
+*.btp.cs
+*.btm.cs
+*.odx.cs
+*.xsd.cs
+
+# OpenCover UI analysis results
+OpenCover/
+
+# Azure Stream Analytics local run output
+ASALocalRun/
+
+# MSBuild Binary and Structured Log
+*.binlog
+
+# NVidia Nsight GPU debugger configuration file
+*.nvuser
+
+# MFractors (Xamarin productivity tool) working folder
+.mfractor/
+
+# Local History for Visual Studio
+.localhistory/
+
+# BeatPulse healthcheck temp database
+healthchecksdb
+
+# Backup folder for Package Reference Convert tool in Visual Studio 2017
+MigrationBackup/
+
+# Ionide (cross platform F# VS Code tools) working folder
+.ionide/
+
+# Fody - auto-generated XML schema
+FodyWeavers.xsd
+
+# build
+build
+monotonic_align/core.c
+*.o
+*.so
+*.dll
+
+# data
+/config.json
+/*.pth
+*.wav
+/monotonic_align/monotonic_align
+/resources
+/MoeGoe.spec
+/dist/MoeGoe
+/dist
+
+.idea

hubert_base.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f54b40fd2802423a5643779c4861af1e9ee9c1564dc9d32f54f20b5ffba7db96
+size 189507909

requirements.txt

@@ -0,0 +1,21 @@
+wheel
+setuptools
+ffmpeg
+numba==0.56.4
+numpy==1.23.5
+scipy==1.9.3
+librosa==0.9.1
+fairseq==0.12.2
+faiss-cpu==1.7.3
+gradio==3.36.1
+pyworld==0.3.2
+soundfile>=0.12.1
+praat-parselmouth>=0.4.2
+httpx==0.23.0
+tensorboard
+tensorboardX
+torchcrepe
+onnxruntime
+demucs
+edge-tts
+yt_dlp

rmvpe.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5ed4719f59085d1affc5d81354c70828c740584f2d24e782523345a6a278962
+size 181189687

rmvpe.py

@@ -0,0 +1,432 @@
+import sys, torch, numpy as np, traceback, pdb
+import torch.nn as nn
+from time import time as ttime
+import torch.nn.functional as F
+
+
+class BiGRU(nn.Module):
+    def __init__(self, input_features, hidden_features, num_layers):
+        super(BiGRU, self).__init__()
+        self.gru = nn.GRU(
+            input_features,
+            hidden_features,
+            num_layers=num_layers,
+            batch_first=True,
+            bidirectional=True,
+        )
+
+    def forward(self, x):
+        return self.gru(x)[0]
+
+
+class ConvBlockRes(nn.Module):
+    def __init__(self, in_channels, out_channels, momentum=0.01):
+        super(ConvBlockRes, self).__init__()
+        self.conv = nn.Sequential(
+            nn.Conv2d(
+                in_channels=in_channels,
+                out_channels=out_channels,
+                kernel_size=(3, 3),
+                stride=(1, 1),
+                padding=(1, 1),
+                bias=False,
+            ),
+            nn.BatchNorm2d(out_channels, momentum=momentum),
+            nn.ReLU(),
+            nn.Conv2d(
+                in_channels=out_channels,
+                out_channels=out_channels,
+                kernel_size=(3, 3),
+                stride=(1, 1),
+                padding=(1, 1),
+                bias=False,
+            ),
+            nn.BatchNorm2d(out_channels, momentum=momentum),
+            nn.ReLU(),
+        )
+        if in_channels != out_channels:
+            self.shortcut = nn.Conv2d(in_channels, out_channels, (1, 1))
+            self.is_shortcut = True
+        else:
+            self.is_shortcut = False
+
+    def forward(self, x):
+        if self.is_shortcut:
+            return self.conv(x) + self.shortcut(x)
+        else:
+            return self.conv(x) + x
+
+
+class Encoder(nn.Module):
+    def __init__(
+        self,
+        in_channels,
+        in_size,
+        n_encoders,
+        kernel_size,
+        n_blocks,
+        out_channels=16,
+        momentum=0.01,
+    ):
+        super(Encoder, self).__init__()
+        self.n_encoders = n_encoders
+        self.bn = nn.BatchNorm2d(in_channels, momentum=momentum)
+        self.layers = nn.ModuleList()
+        self.latent_channels = []
+        for i in range(self.n_encoders):
+            self.layers.append(
+                ResEncoderBlock(
+                    in_channels, out_channels, kernel_size, n_blocks, momentum=momentum
+                )
+            )
+            self.latent_channels.append([out_channels, in_size])
+            in_channels = out_channels
+            out_channels *= 2
+            in_size //= 2
+        self.out_size = in_size
+        self.out_channel = out_channels
+
+    def forward(self, x):
+        concat_tensors = []
+        x = self.bn(x)
+        for i in range(self.n_encoders):
+            _, x = self.layers[i](x)
+            concat_tensors.append(_)
+        return x, concat_tensors
+
+
+class ResEncoderBlock(nn.Module):
+    def __init__(
+        self, in_channels, out_channels, kernel_size, n_blocks=1, momentum=0.01
+    ):
+        super(ResEncoderBlock, self).__init__()
+        self.n_blocks = n_blocks
+        self.conv = nn.ModuleList()
+        self.conv.append(ConvBlockRes(in_channels, out_channels, momentum))
+        for i in range(n_blocks - 1):
+            self.conv.append(ConvBlockRes(out_channels, out_channels, momentum))
+        self.kernel_size = kernel_size
+        if self.kernel_size is not None:
+            self.pool = nn.AvgPool2d(kernel_size=kernel_size)
+
+    def forward(self, x):
+        for i in range(self.n_blocks):
+            x = self.conv[i](x)
+        if self.kernel_size is not None:
+            return x, self.pool(x)
+        else:
+            return x
+
+
+class Intermediate(nn.Module):  #
+    def __init__(self, in_channels, out_channels, n_inters, n_blocks, momentum=0.01):
+        super(Intermediate, self).__init__()
+        self.n_inters = n_inters
+        self.layers = nn.ModuleList()
+        self.layers.append(
+            ResEncoderBlock(in_channels, out_channels, None, n_blocks, momentum)
+        )
+        for i in range(self.n_inters - 1):
+            self.layers.append(
+                ResEncoderBlock(out_channels, out_channels, None, n_blocks, momentum)
+            )
+
+    def forward(self, x):
+        for i in range(self.n_inters):
+            x = self.layers[i](x)
+        return x
+
+
+class ResDecoderBlock(nn.Module):
+    def __init__(self, in_channels, out_channels, stride, n_blocks=1, momentum=0.01):
+        super(ResDecoderBlock, self).__init__()
+        out_padding = (0, 1) if stride == (1, 2) else (1, 1)
+        self.n_blocks = n_blocks
+        self.conv1 = nn.Sequential(
+            nn.ConvTranspose2d(
+                in_channels=in_channels,
+                out_channels=out_channels,
+                kernel_size=(3, 3),
+                stride=stride,
+                padding=(1, 1),
+                output_padding=out_padding,
+                bias=False,
+            ),
+            nn.BatchNorm2d(out_channels, momentum=momentum),
+            nn.ReLU(),
+        )
+        self.conv2 = nn.ModuleList()
+        self.conv2.append(ConvBlockRes(out_channels * 2, out_channels, momentum))
+        for i in range(n_blocks - 1):
+            self.conv2.append(ConvBlockRes(out_channels, out_channels, momentum))
+
+    def forward(self, x, concat_tensor):
+        x = self.conv1(x)
+        x = torch.cat((x, concat_tensor), dim=1)
+        for i in range(self.n_blocks):
+            x = self.conv2[i](x)
+        return x
+
+
+class Decoder(nn.Module):
+    def __init__(self, in_channels, n_decoders, stride, n_blocks, momentum=0.01):
+        super(Decoder, self).__init__()
+        self.layers = nn.ModuleList()
+        self.n_decoders = n_decoders
+        for i in range(self.n_decoders):
+            out_channels = in_channels // 2
+            self.layers.append(
+                ResDecoderBlock(in_channels, out_channels, stride, n_blocks, momentum)
+            )
+            in_channels = out_channels
+
+    def forward(self, x, concat_tensors):
+        for i in range(self.n_decoders):
+            x = self.layers[i](x, concat_tensors[-1 - i])
+        return x
+
+
+class DeepUnet(nn.Module):
+    def __init__(
+        self,
+        kernel_size,
+        n_blocks,
+        en_de_layers=5,
+        inter_layers=4,
+        in_channels=1,
+        en_out_channels=16,
+    ):
+        super(DeepUnet, self).__init__()
+        self.encoder = Encoder(
+            in_channels, 128, en_de_layers, kernel_size, n_blocks, en_out_channels
+        )
+        self.intermediate = Intermediate(
+            self.encoder.out_channel // 2,
+            self.encoder.out_channel,
+            inter_layers,
+            n_blocks,
+        )
+        self.decoder = Decoder(
+            self.encoder.out_channel, en_de_layers, kernel_size, n_blocks
+        )
+
+    def forward(self, x):
+        x, concat_tensors = self.encoder(x)
+        x = self.intermediate(x)
+        x = self.decoder(x, concat_tensors)
+        return x
+
+
+class E2E(nn.Module):
+    def __init__(
+        self,
+        n_blocks,
+        n_gru,
+        kernel_size,
+        en_de_layers=5,
+        inter_layers=4,
+        in_channels=1,
+        en_out_channels=16,
+    ):
+        super(E2E, self).__init__()
+        self.unet = DeepUnet(
+            kernel_size,
+            n_blocks,
+            en_de_layers,
+            inter_layers,
+            in_channels,
+            en_out_channels,
+        )
+        self.cnn = nn.Conv2d(en_out_channels, 3, (3, 3), padding=(1, 1))
+        if n_gru:
+            self.fc = nn.Sequential(
+                BiGRU(3 * 128, 256, n_gru),
+                nn.Linear(512, 360),
+                nn.Dropout(0.25),
+                nn.Sigmoid(),
+            )
+        else:
+            self.fc = nn.Sequential(
+                nn.Linear(3 * N_MELS, N_CLASS), nn.Dropout(0.25), nn.Sigmoid()
+            )
+
+    def forward(self, mel):
+        mel = mel.transpose(-1, -2).unsqueeze(1)
+        x = self.cnn(self.unet(mel)).transpose(1, 2).flatten(-2)
+        x = self.fc(x)
+        return x
+
+
+from librosa.filters import mel
+
+
+class MelSpectrogram(torch.nn.Module):
+    def __init__(
+        self,
+        is_half,
+        n_mel_channels,
+        sampling_rate,
+        win_length,
+        hop_length,
+        n_fft=None,
+        mel_fmin=0,
+        mel_fmax=None,
+        clamp=1e-5,
+    ):
+        super().__init__()
+        n_fft = win_length if n_fft is None else n_fft
+        self.hann_window = {}
+        mel_basis = mel(
+            sr=sampling_rate,
+            n_fft=n_fft,
+            n_mels=n_mel_channels,
+            fmin=mel_fmin,
+            fmax=mel_fmax,
+            htk=True,
+        )
+        mel_basis = torch.from_numpy(mel_basis).float()
+        self.register_buffer("mel_basis", mel_basis)
+        self.n_fft = win_length if n_fft is None else n_fft
+        self.hop_length = hop_length
+        self.win_length = win_length
+        self.sampling_rate = sampling_rate
+        self.n_mel_channels = n_mel_channels
+        self.clamp = clamp
+        self.is_half = is_half
+
+    def forward(self, audio, keyshift=0, speed=1, center=True):
+        factor = 2 ** (keyshift / 12)
+        n_fft_new = int(np.round(self.n_fft * factor))
+        win_length_new = int(np.round(self.win_length * factor))
+        hop_length_new = int(np.round(self.hop_length * speed))
+        keyshift_key = str(keyshift) + "_" + str(audio.device)
+        if keyshift_key not in self.hann_window:
+            self.hann_window[keyshift_key] = torch.hann_window(win_length_new).to(
+                audio.device
+            )
+        fft = torch.stft(
+            audio,
+            n_fft=n_fft_new,
+            hop_length=hop_length_new,
+            win_length=win_length_new,
+            window=self.hann_window[keyshift_key],
+            center=center,
+            return_complex=True,
+        )
+        magnitude = torch.sqrt(fft.real.pow(2) + fft.imag.pow(2))
+        if keyshift != 0:
+            size = self.n_fft // 2 + 1
+            resize = magnitude.size(1)
+            if resize < size:
+                magnitude = F.pad(magnitude, (0, 0, 0, size - resize))
+            magnitude = magnitude[:, :size, :] * self.win_length / win_length_new
+        mel_output = torch.matmul(self.mel_basis, magnitude)
+        if self.is_half == True:
+            mel_output = mel_output.half()
+        log_mel_spec = torch.log(torch.clamp(mel_output, min=self.clamp))
+        return log_mel_spec
+
+
+class RMVPE:
+    def __init__(self, model_path, is_half, device=None):
+        self.resample_kernel = {}
+        model = E2E(4, 1, (2, 2))
+        ckpt = torch.load(model_path, map_location="cpu")
+        model.load_state_dict(ckpt)
+        model.eval()
+        if is_half == True:
+            model = model.half()
+        self.model = model
+        self.resample_kernel = {}
+        self.is_half = is_half
+        if device is None:
+            device = "cuda" if torch.cuda.is_available() else "cpu"
+        self.device = device
+        self.mel_extractor = MelSpectrogram(
+            is_half, 128, 16000, 1024, 160, None, 30, 8000
+        ).to(device)
+        self.model = self.model.to(device)
+        cents_mapping = 20 * np.arange(360) + 1997.3794084376191
+        self.cents_mapping = np.pad(cents_mapping, (4, 4))  # 368
+
+    def mel2hidden(self, mel):
+        with torch.no_grad():
+            n_frames = mel.shape[-1]
+            mel = F.pad(
+                mel, (0, 32 * ((n_frames - 1) // 32 + 1) - n_frames), mode="reflect"
+            )
+            hidden = self.model(mel)
+            return hidden[:, :n_frames]
+
+    def decode(self, hidden, thred=0.03):
+        cents_pred = self.to_local_average_cents(hidden, thred=thred)
+        f0 = 10 * (2 ** (cents_pred / 1200))
+        f0[f0 == 10] = 0
+        # f0 = np.array([10 * (2 ** (cent_pred / 1200)) if cent_pred else 0 for cent_pred in cents_pred])
+        return f0
+
+    def infer_from_audio(self, audio, thred=0.03):
+        audio = torch.from_numpy(audio).float().to(self.device).unsqueeze(0)
+        # torch.cuda.synchronize()
+        # t0=ttime()
+        mel = self.mel_extractor(audio, center=True)
+        # torch.cuda.synchronize()
+        # t1=ttime()
+        hidden = self.mel2hidden(mel)
+        # torch.cuda.synchronize()
+        # t2=ttime()
+        hidden = hidden.squeeze(0).cpu().numpy()
+        if self.is_half == True:
+            hidden = hidden.astype("float32")
+        f0 = self.decode(hidden, thred=thred)
+        # torch.cuda.synchronize()
+        # t3=ttime()
+        # print("hmvpe:%s\t%s\t%s\t%s"%(t1-t0,t2-t1,t3-t2,t3-t0))
+        return f0
+
+    def to_local_average_cents(self, salience, thred=0.05):
+        # t0 = ttime()
+        center = np.argmax(salience, axis=1)  # 帧长#index
+        salience = np.pad(salience, ((0, 0), (4, 4)))  # 帧长,368
+        # t1 = ttime()
+        center += 4
+        todo_salience = []
+        todo_cents_mapping = []
+        starts = center - 4
+        ends = center + 5
+        for idx in range(salience.shape[0]):
+            todo_salience.append(salience[:, starts[idx] : ends[idx]][idx])
+            todo_cents_mapping.append(self.cents_mapping[starts[idx] : ends[idx]])
+        # t2 = ttime()
+        todo_salience = np.array(todo_salience)  # 帧长，9
+        todo_cents_mapping = np.array(todo_cents_mapping)  # 帧长，9
+        product_sum = np.sum(todo_salience * todo_cents_mapping, 1)
+        weight_sum = np.sum(todo_salience, 1)  # 帧长
+        devided = product_sum / weight_sum  # 帧长
+        # t3 = ttime()
+        maxx = np.max(salience, axis=1)  # 帧长
+        devided[maxx <= thred] = 0
+        # t4 = ttime()
+        # print("decode:%s\t%s\t%s\t%s" % (t1 - t0, t2 - t1, t3 - t2, t4 - t3))
+        return devided
+
+
+# if __name__ == '__main__':
+#     audio, sampling_rate = sf.read("卢本伟语录~1.wav")
+#     if len(audio.shape) > 1:
+#         audio = librosa.to_mono(audio.transpose(1, 0))
+#     audio_bak = audio.copy()
+#     if sampling_rate != 16000:
+#         audio = librosa.resample(audio, orig_sr=sampling_rate, target_sr=16000)
+#     model_path = "/bili-coeus/jupyter/jupyterhub-liujing04/vits_ch/test-RMVPE/weights/rmvpe_llc_half.pt"
+#     thred = 0.03  # 0.01
+#     device = 'cuda' if torch.cuda.is_available() else 'cpu'
+#     rmvpe = RMVPE(model_path,is_half=False, device=device)
+#     t0=ttime()
+#     f0 = rmvpe.infer_from_audio(audio, thred=thred)
+#     f0 = rmvpe.infer_from_audio(audio, thred=thred)
+#     f0 = rmvpe.infer_from_audio(audio, thred=thred)
+#     f0 = rmvpe.infer_from_audio(audio, thred=thred)
+#     f0 = rmvpe.infer_from_audio(audio, thred=thred)
+#     t1=ttime()
+#     print(f0.shape,t1-t0)

vc_infer_pipeline.py

@@ -0,0 +1,443 @@
+import numpy as np, parselmouth, torch, pdb, sys, os
+from time import time as ttime
+import torch.nn.functional as F
+import scipy.signal as signal
+import pyworld, os, traceback, faiss, librosa, torchcrepe
+from scipy import signal
+from functools import lru_cache
+
+now_dir = os.getcwd()
+sys.path.append(now_dir)
+
+bh, ah = signal.butter(N=5, Wn=48, btype="high", fs=16000)
+
+input_audio_path2wav = {}
+
+
+@lru_cache
+def cache_harvest_f0(input_audio_path, fs, f0max, f0min, frame_period):
+    audio = input_audio_path2wav[input_audio_path]
+    f0, t = pyworld.harvest(
+        audio,
+        fs=fs,
+        f0_ceil=f0max,
+        f0_floor=f0min,
+        frame_period=frame_period,
+    )
+    f0 = pyworld.stonemask(audio, f0, t, fs)
+    return f0
+
+
+def change_rms(data1, sr1, data2, sr2, rate):  # 1是输入音频，2是输出音频,rate是2的占比
+    # print(data1.max(),data2.max())
+    rms1 = librosa.feature.rms(
+        y=data1, frame_length=sr1 // 2 * 2, hop_length=sr1 // 2
+    )  # 每半秒一个点
+    rms2 = librosa.feature.rms(y=data2, frame_length=sr2 // 2 * 2, hop_length=sr2 // 2)
+    rms1 = torch.from_numpy(rms1)
+    rms1 = F.interpolate(
+        rms1.unsqueeze(0), size=data2.shape[0], mode="linear"
+    ).squeeze()
+    rms2 = torch.from_numpy(rms2)
+    rms2 = F.interpolate(
+        rms2.unsqueeze(0), size=data2.shape[0], mode="linear"
+    ).squeeze()
+    rms2 = torch.max(rms2, torch.zeros_like(rms2) + 1e-6)
+    data2 *= (
+        torch.pow(rms1, torch.tensor(1 - rate))
+        * torch.pow(rms2, torch.tensor(rate - 1))
+    ).numpy()
+    return data2
+
+
+class VC(object):
+    def __init__(self, tgt_sr, config):
+        self.x_pad, self.x_query, self.x_center, self.x_max, self.is_half = (
+            config.x_pad,
+            config.x_query,
+            config.x_center,
+            config.x_max,
+            config.is_half,
+        )
+        self.sr = 16000  # hubert输入采样率
+        self.window = 160  # 每帧点数
+        self.t_pad = self.sr * self.x_pad  # 每条前后pad时间
+        self.t_pad_tgt = tgt_sr * self.x_pad
+        self.t_pad2 = self.t_pad * 2
+        self.t_query = self.sr * self.x_query  # 查询切点前后查询时间
+        self.t_center = self.sr * self.x_center  # 查询切点位置
+        self.t_max = self.sr * self.x_max  # 免查询时长阈值
+        self.device = config.device
+
+    def get_f0(
+        self,
+        input_audio_path,
+        x,
+        p_len,
+        f0_up_key,
+        f0_method,
+        filter_radius,
+        inp_f0=None,
+    ):
+        global input_audio_path2wav
+        time_step = self.window / self.sr * 1000
+        f0_min = 50
+        f0_max = 1100
+        f0_mel_min = 1127 * np.log(1 + f0_min / 700)
+        f0_mel_max = 1127 * np.log(1 + f0_max / 700)
+        if f0_method == "pm":
+            f0 = (
+                parselmouth.Sound(x, self.sr)
+                .to_pitch_ac(
+                    time_step=time_step / 1000,
+                    voicing_threshold=0.6,
+                    pitch_floor=f0_min,
+                    pitch_ceiling=f0_max,
+                )
+                .selected_array["frequency"]
+            )
+            pad_size = (p_len - len(f0) + 1) // 2
+            if pad_size > 0 or p_len - len(f0) - pad_size > 0:
+                f0 = np.pad(
+                    f0, [[pad_size, p_len - len(f0) - pad_size]], mode="constant"
+                )
+        elif f0_method == "harvest":
+            input_audio_path2wav[input_audio_path] = x.astype(np.double)
+            f0 = cache_harvest_f0(input_audio_path, self.sr, f0_max, f0_min, 10)
+            if filter_radius > 2:
+                f0 = signal.medfilt(f0, 3)
+        elif f0_method == "crepe":
+            model = "full"
+            # Pick a batch size that doesn't cause memory errors on your gpu
+            batch_size = 512
+            # Compute pitch using first gpu
+            audio = torch.tensor(np.copy(x))[None].float()
+            f0, pd = torchcrepe.predict(
+                audio,
+                self.sr,
+                self.window,
+                f0_min,
+                f0_max,
+                model,
+                batch_size=batch_size,
+                device=self.device,
+                return_periodicity=True,
+            )
+            pd = torchcrepe.filter.median(pd, 3)
+            f0 = torchcrepe.filter.mean(f0, 3)
+            f0[pd < 0.1] = 0
+            f0 = f0[0].cpu().numpy()
+        elif f0_method == "rmvpe":
+            if hasattr(self, "model_rmvpe") == False:
+                from rmvpe import RMVPE
+
+                print("loading rmvpe model")
+                self.model_rmvpe = RMVPE(
+                    "rmvpe.pt", is_half=self.is_half, device=self.device
+                )
+            f0 = self.model_rmvpe.infer_from_audio(x, thred=0.03)
+        f0 *= pow(2, f0_up_key / 12)
+        # with open("test.txt","w")as f:f.write("\n".join([str(i)for i in f0.tolist()]))
+        tf0 = self.sr // self.window  # 每秒f0点数
+        if inp_f0 is not None:
+            delta_t = np.round(
+                (inp_f0[:, 0].max() - inp_f0[:, 0].min()) * tf0 + 1
+            ).astype("int16")
+            replace_f0 = np.interp(
+                list(range(delta_t)), inp_f0[:, 0] * 100, inp_f0[:, 1]
+            )
+            shape = f0[self.x_pad * tf0 : self.x_pad * tf0 + len(replace_f0)].shape[0]
+            f0[self.x_pad * tf0 : self.x_pad * tf0 + len(replace_f0)] = replace_f0[
+                :shape
+            ]
+        # with open("test_opt.txt","w")as f:f.write("\n".join([str(i)for i in f0.tolist()]))
+        f0bak = f0.copy()
+        f0_mel = 1127 * np.log(1 + f0 / 700)
+        f0_mel[f0_mel > 0] = (f0_mel[f0_mel > 0] - f0_mel_min) * 254 / (
+            f0_mel_max - f0_mel_min
+        ) + 1
+        f0_mel[f0_mel <= 1] = 1
+        f0_mel[f0_mel > 255] = 255
+        f0_coarse = np.rint(f0_mel).astype(np.int)
+        return f0_coarse, f0bak  # 1-0
+
+    def vc(
+        self,
+        model,
+        net_g,
+        sid,
+        audio0,
+        pitch,
+        pitchf,
+        times,
+        index,
+        big_npy,
+        index_rate,
+        version,
+        protect,
+    ):  # ,file_index,file_big_npy
+        feats = torch.from_numpy(audio0)
+        if self.is_half:
+            feats = feats.half()
+        else:
+            feats = feats.float()
+        if feats.dim() == 2:  # double channels
+            feats = feats.mean(-1)
+        assert feats.dim() == 1, feats.dim()
+        feats = feats.view(1, -1)
+        padding_mask = torch.BoolTensor(feats.shape).to(self.device).fill_(False)
+
+        inputs = {
+            "source": feats.to(self.device),
+            "padding_mask": padding_mask,
+            "output_layer": 9 if version == "v1" else 12,
+        }
+        t0 = ttime()
+        with torch.no_grad():
+            logits = model.extract_features(**inputs)
+            feats = model.final_proj(logits[0]) if version == "v1" else logits[0]
+        if protect < 0.5 and pitch != None and pitchf != None:
+            feats0 = feats.clone()
+        if (
+            isinstance(index, type(None)) == False
+            and isinstance(big_npy, type(None)) == False
+            and index_rate != 0
+        ):
+            npy = feats[0].cpu().numpy()
+            if self.is_half:
+                npy = npy.astype("float32")
+
+            # _, I = index.search(npy, 1)
+            # npy = big_npy[I.squeeze()]
+
+            score, ix = index.search(npy, k=8)
+            weight = np.square(1 / score)
+            weight /= weight.sum(axis=1, keepdims=True)
+            npy = np.sum(big_npy[ix] * np.expand_dims(weight, axis=2), axis=1)
+
+            if self.is_half:
+                npy = npy.astype("float16")
+            feats = (
+                torch.from_numpy(npy).unsqueeze(0).to(self.device) * index_rate
+                + (1 - index_rate) * feats
+            )
+
+        feats = F.interpolate(feats.permute(0, 2, 1), scale_factor=2).permute(0, 2, 1)
+        if protect < 0.5 and pitch != None and pitchf != None:
+            feats0 = F.interpolate(feats0.permute(0, 2, 1), scale_factor=2).permute(
+                0, 2, 1
+            )
+        t1 = ttime()
+        p_len = audio0.shape[0] // self.window
+        if feats.shape[1] < p_len:
+            p_len = feats.shape[1]
+            if pitch != None and pitchf != None:
+                pitch = pitch[:, :p_len]
+                pitchf = pitchf[:, :p_len]
+
+        if protect < 0.5 and pitch != None and pitchf != None:
+            pitchff = pitchf.clone()
+            pitchff[pitchf > 0] = 1
+            pitchff[pitchf < 1] = protect
+            pitchff = pitchff.unsqueeze(-1)
+            feats = feats * pitchff + feats0 * (1 - pitchff)
+            feats = feats.to(feats0.dtype)
+        p_len = torch.tensor([p_len], device=self.device).long()
+        with torch.no_grad():
+            if pitch != None and pitchf != None:
+                audio1 = (
+                    (net_g.infer(feats, p_len, pitch, pitchf, sid)[0][0, 0])
+                    .data.cpu()
+                    .float()
+                    .numpy()
+                )
+            else:
+                audio1 = (
+                    (net_g.infer(feats, p_len, sid)[0][0, 0]).data.cpu().float().numpy()
+                )
+        del feats, p_len, padding_mask
+        if torch.cuda.is_available():
+            torch.cuda.empty_cache()
+        t2 = ttime()
+        times[0] += t1 - t0
+        times[2] += t2 - t1
+        return audio1
+
+    def pipeline(
+        self,
+        model,
+        net_g,
+        sid,
+        audio,
+        input_audio_path,
+        times,
+        f0_up_key,
+        f0_method,
+        file_index,
+        # file_big_npy,
+        index_rate,
+        if_f0,
+        filter_radius,
+        tgt_sr,
+        resample_sr,
+        rms_mix_rate,
+        version,
+        protect,
+        f0_file=None,
+    ):
+        if (
+            file_index != ""
+            # and file_big_npy != ""
+            # and os.path.exists(file_big_npy) == True
+            and os.path.exists(file_index) == True
+            and index_rate != 0
+        ):
+            try:
+                index = faiss.read_index(file_index)
+                # big_npy = np.load(file_big_npy)
+                big_npy = index.reconstruct_n(0, index.ntotal)
+            except:
+                traceback.print_exc()
+                index = big_npy = None
+        else:
+            index = big_npy = None
+        audio = signal.filtfilt(bh, ah, audio)
+        audio_pad = np.pad(audio, (self.window // 2, self.window // 2), mode="reflect")
+        opt_ts = []
+        if audio_pad.shape[0] > self.t_max:
+            audio_sum = np.zeros_like(audio)
+            for i in range(self.window):
+                audio_sum += audio_pad[i : i - self.window]
+            for t in range(self.t_center, audio.shape[0], self.t_center):
+                opt_ts.append(
+                    t
+                    - self.t_query
+                    + np.where(
+                        np.abs(audio_sum[t - self.t_query : t + self.t_query])
+                        == np.abs(audio_sum[t - self.t_query : t + self.t_query]).min()
+                    )[0][0]
+                )
+        s = 0
+        audio_opt = []
+        t = None
+        t1 = ttime()
+        audio_pad = np.pad(audio, (self.t_pad, self.t_pad), mode="reflect")
+        p_len = audio_pad.shape[0] // self.window
+        inp_f0 = None
+        if hasattr(f0_file, "name") == True:
+            try:
+                with open(f0_file.name, "r") as f:
+                    lines = f.read().strip("\n").split("\n")
+                inp_f0 = []
+                for line in lines:
+                    inp_f0.append([float(i) for i in line.split(",")])
+                inp_f0 = np.array(inp_f0, dtype="float32")
+            except:
+                traceback.print_exc()
+        sid = torch.tensor(sid, device=self.device).unsqueeze(0).long()
+        pitch, pitchf = None, None
+        if if_f0 == 1:
+            pitch, pitchf = self.get_f0(
+                input_audio_path,
+                audio_pad,
+                p_len,
+                f0_up_key,
+                f0_method,
+                filter_radius,
+                inp_f0,
+            )
+            pitch = pitch[:p_len]
+            pitchf = pitchf[:p_len]
+            if self.device == "mps":
+                pitchf = pitchf.astype(np.float32)
+            pitch = torch.tensor(pitch, device=self.device).unsqueeze(0).long()
+            pitchf = torch.tensor(pitchf, device=self.device).unsqueeze(0).float()
+        t2 = ttime()
+        times[1] += t2 - t1
+        for t in opt_ts:
+            t = t // self.window * self.window
+            if if_f0 == 1:
+                audio_opt.append(
+                    self.vc(
+                        model,
+                        net_g,
+                        sid,
+                        audio_pad[s : t + self.t_pad2 + self.window],
+                        pitch[:, s // self.window : (t + self.t_pad2) // self.window],
+                        pitchf[:, s // self.window : (t + self.t_pad2) // self.window],
+                        times,
+                        index,
+                        big_npy,
+                        index_rate,
+                        version,
+                        protect,
+                    )[self.t_pad_tgt : -self.t_pad_tgt]
+                )
+            else:
+                audio_opt.append(
+                    self.vc(
+                        model,
+                        net_g,
+                        sid,
+                        audio_pad[s : t + self.t_pad2 + self.window],
+                        None,
+                        None,
+                        times,
+                        index,
+                        big_npy,
+                        index_rate,
+                        version,
+                        protect,
+                    )[self.t_pad_tgt : -self.t_pad_tgt]
+                )
+            s = t
+        if if_f0 == 1:
+            audio_opt.append(
+                self.vc(
+                    model,
+                    net_g,
+                    sid,
+                    audio_pad[t:],
+                    pitch[:, t // self.window :] if t is not None else pitch,
+                    pitchf[:, t // self.window :] if t is not None else pitchf,
+                    times,
+                    index,
+                    big_npy,
+                    index_rate,
+                    version,
+                    protect,
+                )[self.t_pad_tgt : -self.t_pad_tgt]
+            )
+        else:
+            audio_opt.append(
+                self.vc(
+                    model,
+                    net_g,
+                    sid,
+                    audio_pad[t:],
+                    None,
+                    None,
+                    times,
+                    index,
+                    big_npy,
+                    index_rate,
+                    version,
+                    protect,
+                )[self.t_pad_tgt : -self.t_pad_tgt]
+            )
+        audio_opt = np.concatenate(audio_opt)
+        if rms_mix_rate != 1:
+            audio_opt = change_rms(audio, 16000, audio_opt, tgt_sr, rms_mix_rate)
+        if resample_sr >= 16000 and tgt_sr != resample_sr:
+            audio_opt = librosa.resample(
+                audio_opt, orig_sr=tgt_sr, target_sr=resample_sr
+            )
+        audio_max = np.abs(audio_opt).max() / 0.99
+        max_int16 = 32768
+        if audio_max > 1:
+            max_int16 /= audio_max
+        audio_opt = (audio_opt * max_int16).astype(np.int16)
+        del pitch, pitchf, sid
+        if torch.cuda.is_available():
+            torch.cuda.empty_cache()
+        return audio_opt