app.py

from transformers import pipeline, AutoModelForSpeechSeq2Seq, AutoProcessor
from transformers.pipelines.audio_utils import ffmpeg_read
from huggingface_hub import login
import yt_dlp as youtube_dl
import gradio as gr
import tempfile
import spaces
import torch
import time
import os

login(os.environ["HF"], add_to_git_credential=True)

BATCH_SIZE = 16
FILE_LIMIT_MB = 1000
YT_LENGTH_LIMIT_S = 3600  # limit to 1 hour YouTube files

device = "cuda:0" if torch.cuda.is_available() else "cpu"
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
model_id = "Kushtrim/whisper-large-v3-turbo-shqip"
model = AutoModelForSpeechSeq2Seq.from_pretrained(model_id, torch_dtype=torch_dtype, use_safetensors=True).to(device)
processor = AutoProcessor.from_pretrained(model_id)
pipe = pipeline("automatic-speech-recognition", model=model, tokenizer=processor.tokenizer, feature_extractor=processor.feature_extractor,
                max_new_tokens=256, chunk_length_s=28, batch_size=16, torch_dtype=torch_dtype, device=device,
                token=os.environ["HF"])

@spaces.GPU
def transcribe(inputs, task):
    if inputs is None:
        raise gr.Error(
            "No audio file submitted! Please upload or record an audio file before submitting your request.")

    text = pipe(inputs, generate_kwargs={
                'num_beams': 5, "task": task, 'language': 'sq'}, return_timestamps=True)["text"]
    return 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 download_yt_audio(yt_url, filename):
    info_loader = youtube_dl.YoutubeDL()

    try:
        info = info_loader.extract_info(yt_url, download=False)
    except youtube_dl.utils.DownloadError as err:
        raise gr.Error(str(err))

    file_length = info["duration_string"]
    file_h_m_s = file_length.split(":")
    file_h_m_s = [int(sub_length) for sub_length in file_h_m_s]

    if len(file_h_m_s) == 1:
        file_h_m_s.insert(0, 0)
    if len(file_h_m_s) == 2:
        file_h_m_s.insert(0, 0)
    file_length_s = file_h_m_s[0] * 3600 + file_h_m_s[1] * 60 + file_h_m_s[2]

    if file_length_s > YT_LENGTH_LIMIT_S:
        yt_length_limit_hms = time.strftime(
            "%HH:%MM:%SS", time.gmtime(YT_LENGTH_LIMIT_S))
        file_length_hms = time.strftime(
            "%HH:%MM:%SS", time.gmtime(file_length_s))
        raise gr.Error(f"Maximum YouTube length is {yt_length_limit_hms}, got {file_length_hms} YouTube video.")

    ydl_opts = {"outtmpl": filename,
                "format": "worstvideo[ext=mp4]+bestaudio[ext=m4a]/best[ext=mp4]/best"}

    with youtube_dl.YoutubeDL(ydl_opts) as ydl:
        try:
            ydl.download([yt_url])
        except youtube_dl.utils.ExtractorError as err:
            raise gr.Error(str(err))


def yt_transcribe(yt_url, task, max_filesize=75.0):
    html_embed_str = _return_yt_html_embed(yt_url)

    with tempfile.TemporaryDirectory() as tmpdirname:
        filepath = os.path.join(tmpdirname, "video.mp4")
        download_yt_audio(yt_url, filepath)
        with open(filepath, "rb") as f:
            inputs = f.read()

    inputs = ffmpeg_read(inputs, pipe.feature_extractor.sampling_rate)
    inputs = {"array": inputs,
              "sampling_rate": pipe.feature_extractor.sampling_rate}

    text = pipe(inputs, batch_size=BATCH_SIZE, generate_kwargs={
                "task": task}, return_timestamps=True)["text"]

    return html_embed_str, text


demo = gr.Blocks()

file_transcribe = gr.Interface(
    fn=transcribe,
    inputs=[
        gr.Audio(sources=["upload"], type="filepath", label="Audio file"),
    ],
    outputs="text",
    title="Whisper Large V3 Turbo Shqip: Transcribe Audio",
    description=("This fine-tuned Whisper model provides reliable transcription for Albanian audio, whether from a microphone or an uploaded file. " 
                 "Key details about this project:"
                 "\n\n- Fine-tuned on 200 hours of carefully curated Albanian audio data. "
                 "\n- This is the third training run, reflecting continuous improvements as the dataset evolves. "
                 f"\n- Hosted on Hugging Face. Repository: [{model_id}](https://huggingface.co/{model_id}). "
                ),
    allow_flagging="never",
)

mf_transcribe = gr.Interface(
    fn=transcribe,
    inputs=[
        gr.Audio(sources=["microphone"], type="filepath"),
    ],
    outputs="text",
    title="Whisper Large V3 Turbo Shqip: Transcribe Audio",
    description=("This fine-tuned Whisper model provides reliable transcription for Albanian audio, whether from a microphone or an uploaded file. " 
                 "Key details about this project:"
                 "\n\n- Fine-tuned on 200 hours of carefully curated Albanian audio data. "
                 "\n- This is the third training run, reflecting continuous improvements as the dataset evolves. "
                 f"\n- Hosted on Hugging Face. Repository: [{model_id}](https://huggingface.co/{model_id}). "
                ),
    allow_flagging="never",
)

yt_transcribe = gr.Interface(
    fn=yt_transcribe,
    inputs=[
        gr.Textbox(
            lines=1, placeholder="Paste the URL to a YouTube video here", label="YouTube URL"),
    ],
    outputs=["html", "text"],
    title="Whisper Large V3 Turbo Shqip: Transcribe Audio",
    description=("This fine-tuned Whisper model provides reliable transcription for Albanian audio, whether from a microphone or an uploaded file. " 
                 "Key details about this project:"
                 "\n\n- Fine-tuned on 200 hours of carefully curated Albanian audio data. "
                 "\n- This is the third training run, reflecting continuous improvements as the dataset evolves. "
                 f"\n- Hosted on Hugging Face. Repository: [{model_id}](https://huggingface.co/{model_id}). "
                ),
    allow_flagging="never",
)

with demo:
    gr.TabbedInterface([mf_transcribe, file_transcribe, yt_transcribe], ["Microphone", "Audio file", "YouTube"])

demo.launch()