mp3_to_json.py

import whisper
import json
import os
import torch

# model = whisper.load_model("large-v2")
device = "cuda" if torch.cuda.is_available() else "cpu"

model = whisper.load_model("medium", device = device)
print(f"Model loaded successfully on", device)

audios = os.listdir("audios")

for audio in audios:
    if ("_" in audio):
        number = audio.split("_")[0]
        title = audio.split("_")[1][:-4]
        json_name = audio[:-4]

        result = model.transcribe(
            audio = f"audios/{audio}", 
            # audio = "audios/sample.mp3", 
            language="hi",
            task="translate",
            word_timestamps = False
            )
        
        chunks = []
        for segment in result["segments"]:
            chunks.append({
                "number": number,
                "title": title,
                "start": segment["start"], 
                "end": segment["end"], 
                "text": segment["text"]
                })
            
        chunks_with_metadata = {"chunks": chunks, "text": result["text"]}
            
        with open(f"jsons/{json_name}.json", "w") as f:
            json.dump(chunks_with_metadata, f)

    print(f"Audio Number {number} Converted in json Successfully!!!")