app.py

from transformers import pipeline
import gradio as gr
import os
import pytube as pt
import subprocess
import sys

youtube_file_path = "youtube_audio"

pipe = pipeline(model="ArtificialCoincidence/check_points")  # change to "your-username/the-name-you-picked"

def transcribe(audio):
    text = pipe(audio)["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 transcribe_youtube(yt_url):
    yt = pt.YouTube(yt_url)
    html_embed_str = _return_yt_html_embed(yt_url)
    stream = yt.streams.filter(only_audio=True)[0]
    stream.download(filename=youtube_file_path)
    text = transcribe(youtube_file_path)
    return text

def convert_video_to_audio_ffmpeg(video_file, output_ext="mp3"):
    filename, ext = os.path.splitext(video_file)
    subprocess.call(["ffmpeg", "-y", "-i", video_file, f"{filename}.{output_ext}"], 
                    stdout=subprocess.DEVNULL,
                    stderr=subprocess.STDOUT)
    return f"{filename}.{output_ext}"

def transcribe_video(video):
    audio_filename = convert_video_to_audio_ffmpeg(video)
    text = transcribe(audio_filename)
    return text

iface = gr.Blocks()

microphone_trans = gr.Interface(
    fn=transcribe,
    inputs=gr.Audio(sources="microphone", type="filepath"),
    outputs="text",
    title="Whisper Chinese",
    description="Realtime demo for Chinese speech recognition using a fine-tuned Whisper small model.",
)

youtube_trans = gr.Interface(
    fn=transcribe_youtube,
    inputs=gr.Textbox(lines=1, placeholder="Paste the URL to a video here", label="video url"),
    outputs="text",
    title="Whisper Chinese",
    description="Realtime demo for Chinese speech recognition using a fine-tuned Whisper small model.",
)

video_trans = gr.Interface(
    fn=transcribe_video,
    inputs=gr.Video(sources="upload"),
    outputs="text",
    title="Whisper Chinese",
    description="Realtime demo for Chinese speech recognition using a fine-tuned Whisper small model.",
)

with iface:
    gr.TabbedInterface([microphone_trans, youtube_trans, video_trans], ["Transcribe Microphone", "Transcribe Youtube Video", "Transcribe Video"])
iface.launch(debug=True)