File size: 3,110 Bytes
91e1c8c
 
 
 
 
 
 
5e5e671
91e1c8c
 
206de53
91e1c8c
 
 
 
 
5e5e671
61616cc
5e5e671
 
 
61616cc
5e5e671
61616cc
 
5e5e671
61616cc
5e5e671
 
61616cc
5e5e671
9acdcb9
9cc12c0
91e1c8c
 
5e5e671
61616cc
 
 
 
 
9cc12c0
91e1c8c
5e5e671
91e1c8c
5e5e671
91e1c8c
 
 
5e5e671
91e1c8c
 
 
 
 
 
de72597
91e1c8c
 
b113478
 
91e1c8c
b113478
 
 
91e1c8c
b113478
 
 
 
 
 
 
91e1c8c
b113478
 
 
 
 
61616cc
bd941fe
b113478
 
 
bd941fe
b113478
9cc12c0
 
b113478
 
bd941fe
b113478
 
bd941fe
91e1c8c
 
 
 
 
 
 
 
a4a75eb
91e1c8c
 
a4a75eb
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
import tempfile
from typing import Optional
from TTS.config import load_config
import gradio as gr
import numpy as np
from TTS.utils.manage import ModelManager
from TTS.utils.synthesizer import Synthesizer
import os


MODEL_NAMES = ["Celtia"]#, "Icia", "Sabela"]

# reorder models
print(MODEL_NAMES)


def tts(text: str,
        choosen_model: str = "sabela.pth"
):

    # if text is "celtia" take celtia.pth and celtia_config.json
    if choosen_model == "Celtia":
        config_path = "celtia_config.json"
        model_file = "celtia.pth"
    elif choosen_model == "Icia":
        config_path = "icia_config.json"
        model_file = "icia.pth"
    else:
        config_path = "sabela_config.json"
        model_file = "sabela.pth"

    model_path = os.path.join(os.getcwd(), model_file)

    vocoder_path = None
    vocoder_config_path = None

    print(f"Using model: {model_path}\n"
        f"Using config: {config_path}\n"
        f"Using vocoder: {vocoder_path}\n"
        f"Using vocoder config: {vocoder_config_path}"
        )

    synthesizer = Synthesizer(
        model_path, config_path, None, vocoder_path, vocoder_config_path,
    )

    # synthesize
    if synthesizer is None:
        raise NameError("model not found")
    wavs = synthesizer.tts(text)
    # return output
    with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as fp:
        synthesizer.save_wav(wavs, fp)
        return fp.name


title = """<h1 align="center">馃惛馃挰 CoquiTTS Demo Proxecto N贸s </h1>"""

with gr.Blocks(analytics_enabled=False) as demo:
    with gr.Row():
        with gr.Column():
            gr.Markdown(
                    """
                    ## <img src="https://huggingface.co/spaces/proxectonos/README/resolve/main/title-card.png" width="100%" style="border-radius: 0.75rem;">
                    """
            )
        with gr.Column():
            with gr.Row():            
                gr.Markdown(
                """
                <br/>
                
                馃捇 Este space mostra alg煤ns dos modelos TTS desenvolvidos polo **[Proxecto N贸s](https://huggingface.co/proxectonos)**.

                <br/>
                """
                )
            with gr.Row():            
                input_text = gr.Textbox(
                    label="Introduce un texto de entrada:",
                    value="Probando unha frase nova.",
                )
            with gr.Row():            
                model_select = gr.Dropdown(
                    label="Escolle un modelo:",
                    choices=MODEL_NAMES,
                    value=MODEL_NAMES[0],
                    interactive=True
                )
            with gr.Row():            
                tts_button = gr.Button("Enviar", elem_id="send-btn", visible=True)

            with gr.Row():
                output_audio = gr.Audio(label="Sa铆da", type="filepath")

    tts_button.click(
        tts,
        inputs=[
            input_text,
            model_select,
        ],
        outputs=[output_audio],
        concurrency_limit=16,
    )

demo.launch(debug=True)