app.py

import gradio as gr
import openai
import youtube_dl


def asr(url):
    # download audio
    # Options for youtube-dl
    ydl_opts = {
        'outtmpl': 'my_video.mp4'
    }

    # Create a youtube-dl object
    ydl = youtube_dl.YoutubeDL(ydl_opts)

    # Download the video
    info_dict = ydl.extract_info(url, download=True)

    audio_file= open("my_video.mp4", "rb")
    transcript = openai.Audio.transcribe("whisper-1", audio_file)
    output = openai.ChatCompletion.create(
    model="gpt-3.5-turbo",
    messages=[
            {"role": "user", "content": "Transcript: {transcript}. \n Translate the video conversation transcript into fluent Chinese. Chinese: ".format(transcript=transcript["text"])},
        ]
    )
    return transcript["text"], output['choices'][0]['message']['content']

demo = gr.Interface(fn=asr, inputs="text", 
                    outputs=[
                        gr.outputs.Textbox(label="English"),
                        gr.outputs.Textbox(label="Chinese"),
                    ])

demo.launch()