app.py

import time
import os
import re

import torch

import gradio as gr
import spaces
from transformers import AutoFeatureExtractor, AutoTokenizer, WhisperForConditionalGeneration, WhisperProcessor, pipeline
from huggingface_hub import model_info
try:
    import flash_attn
    FLASH_ATTENTION = True
except ImportError:
    FLASH_ATTENTION = False

import yt_dlp  # Added import for yt-dlp

MODEL_NAME = "NbAiLab/nb-whisper-large"
lang = "no"

logo_path = os.path.join(os.path.dirname(__file__), "Logo.png")

share = (os.environ.get("SHARE", "False")[0].lower() in "ty1") or None
auth_token = os.environ.get("AUTH_TOKEN") or True
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print(f"Bruker enhet: {device}")

@spaces.GPU(duration=60 * 2)
def pipe(file, return_timestamps=False):
    asr = pipeline(
        task="automatic-speech-recognition",
        model=MODEL_NAME,
        chunk_length_s=28,
        device=device,
        token=auth_token,
        torch_dtype=torch.float16,
        model_kwargs={"attn_implementation": "flash_attention_2", "num_beams": 5} if FLASH_ATTENTION else {"attn_implementation": "sdpa", "num_beams": 5},
    )
    asr.model.config.forced_decoder_ids = asr.tokenizer.get_decoder_prompt_ids(
        language=lang,
        task="transcribe",
        no_timestamps=not return_timestamps,
    )
    return asr(file, return_timestamps=return_timestamps, batch_size=24)

def format_output(text):
    # Add a newline after ".", "!", ":", or "?" unless part of sequences like "..."
    text = re.sub(r'(?<!\.)[.!:?](?!\.)', lambda m: m.group() + '\n', text)
    # Ensure newline after sequences like "..." or other punctuation patterns
    text = re.sub(r'(\.{3,}|[.!:?])', lambda m: m.group() + '\n\n', text)
    return text

def transcribe(file, return_timestamps=False):
    if not return_timestamps:
        text = pipe(file)["text"]
        formatted_text = format_output(text)
    else:
        chunks = pipe(file, return_timestamps=True)["chunks"]
        text = []
        for chunk in chunks:
            start_time = time.strftime('%H:%M:%S', time.gmtime(chunk["timestamp"][0])) if chunk["timestamp"][0] is not None else "??:??:??"
            end_time = time.strftime('%H:%M:%S', time.gmtime(chunk["timestamp"][1])) if chunk["timestamp"][1] is not None else "??:??:??"
            line = f"[{start_time} -> {end_time}] {chunk['text']}"
            text.append(line)
        formatted_text = "\n".join(text)
    formatted_text += "<br><br><i>Transkribert med NB-Whisper demo</i>"
    return formatted_text

def _return_yt_html_embed(yt_url):
    video_id = yt_url.split("?v=")[-1]
    HTML_str = (
        f'<center> <iframe width="500" height="320" src="https://www.youtube.com/embed/{video_id}"> </iframe>'
        " </center>"
    )
    return HTML_str

def yt_transcribe(yt_url, return_timestamps=False):
    html_embed_str = _return_yt_html_embed(yt_url)

    ydl_opts = {
        'format': 'bestaudio/best',
        'outtmpl': 'audio.%(ext)s',
        'postprocessors': [{
            'key': 'FFmpegExtractAudio',
            'preferredcodec': 'mp3',
            'preferredquality': '192',
        }],
        'quiet': True,
    }
    with yt_dlp.YoutubeDL(ydl_opts) as ydl:
        ydl.download([yt_url])

    text = transcribe("audio.mp3", return_timestamps=return_timestamps)

    return html_embed_str, text

# Lag Gradio-appen uten faner

demo = gr.Blocks()

with demo:
    gr.Image(value=logo_path, type="filepath", label=None, width=100)
    mf_transcribe = gr.Interface(
        fn=transcribe,
        inputs=[
            gr.components.Audio(sources=['upload', 'microphone'], type="filepath"),
            gr.components.Checkbox(label="Inkluder tidsstempler"),
        ],
        outputs=gr.HTML(label="text"),
        title="NB-Whisper",
        description=(
            "Transkriber lange lydopptak fra mikrofon eller lydfiler med et enkelt klikk! Demoen bruker den fintunede"
            f" modellen [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) og 🤗 Transformers til å transkribere lydfiler opp til 30 minutter."
        ),
        allow_flagging="never",
        #show_submit_button=False,
    )

    # Uncomment to add the YouTube transcription interface if needed
    # yt_transcribe_interface = gr.Interface(
    #     fn=yt_transcribe,
    #     inputs=[
    #         gr.components.Textbox(lines=1, placeholder="Lim inn URL til en YouTube-video her", label="YouTube URL"),
    #         gr.components.Checkbox(label="Inkluder tidsstempler"),
    #     ],
    #     examples=[["https://www.youtube.com/watch?v=mukeSSa5GKo"]],
    #     outputs=["html", "text"],
    #     title="Whisper Demo: Transkriber YouTube",
    #     description=(
    #         "Transkriber lange YouTube-videoer med et enkelt klikk! Demoen bruker den fintunede modellen:"
    #         f" [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) og 🤗 Transformers til å transkribere lydfiler av"
    #         " vilkårlig lengde."
    #     ),
    #     allow_flagging="never",
    # )

# Start demoen uten faner
demo.launch(share=share, show_api=False).queue()