app.py

import json
import ffmpeg
from subprocess import run
import gradio as gr
import whisper_timestamped as whisper
from transformers import pipeline

model = whisper.load_model("model/small.pt", device="cpu")
sentiment_analysis = pipeline("sentiment-analysis", framework="pt", model="SamLowe/roberta-base-go_emotions", use_fast=True)

def analyze_sentiment(text):
    results = sentiment_analysis(text)
    sentiment_results = {result['label']: result['score'] for result in results}
    return sentiment_results


def transcribe(audio):
    audio = whisper.load_audio(audio)
    result = whisper.transcribe(model, audio)
    print(json.dumps(result, indent=2, ensure_ascii=False))
    sent_res = analyze_sentiment(result.text)

    return sent_res


def video_to_audio(input_video, output_audio):
    audio_file = f"test.wav"
    run(["ffmpeg", "-i", 'test_video_1.mp4', audio_file])

    response = transcribe(audio=audio_file)

    return response

gr.Interface(
    fn=video_to_audio,
    inputs=gr.Video(),
    outputs=gr.Textbox()
).launch()