Transcribe an audio/video file with Groq Whisper

whisper_groq2srt_shorter.py

from groq import Groq
from datetime import timedelta

def format_time(seconds):
        # Ensure the first segment starts at 00:00:00,000
    if float(seconds) < 0.001:
        return "00:00:00,000"
    
    time = str(timedelta(seconds=float(seconds))).replace('.', ',')[:-3]
    return f"{time:0>12}"

filename = "bbc_ai_edit.mp3"
with open(filename, "rb") as file:
    segments = Groq().audio.transcriptions.create(
        file=(filename, file.read()),
        model="whisper-large-v3-turbo",
        response_format="verbose_json"
    ).segments

with open("subs.srt", 'w') as f:
    f.writelines(f"{i+1}\n{format_time(s['start'])} --> {format_time(s['end'])}\n{s['text'].strip()}\n\n" 
                for i, s in enumerate(segments))

print("SRT file saved successfully!")