Add audio transcription script using Faster Whisper

[Syed-56]

PR Title

Add audio transcription script using Faster Whisper

Summary

Added a new Python script that transcribes audio and video files (MP3, WAV, MP4, etc.) into text using the Faster Whisper library. The script automatically installs the library if missing and checks for ffmpeg.

Description

This script allows users to transcribe audio files quickly and efficiently. It supports multiple model sizes (tiny, small, base, medium, large) and allows English-only or multilingual transcription. The script validates the input file, prompts for model choice, and outputs the transcript in the same folder.

The changes are as follows:

• Added audio_transcribe.py script
• Auto-installs faster_whisper if missing
• Checks for ffmpeg and provides installation instructions
• Supports MP3, WAV, M4A, FLAC, AAC, and MP4 files
• Prompts user for model size and English-only/multilingual choice
• Saves transcription output as <audio_file>_transcript.txt
• Includes user-friendly messages and error handling

Checks

in the repository

• Made no changes that degrade the functioning of the repository
• Gave each commit a descriptive title

in the PR

• Followed the format of the pull_request_template
• Made the Pull Request concise and self-contained
• Tested the script locally and verified transcription works

Thank You,

Syed Sultan

[Syed-56]

Validation and loop made for both model and file. Also some minor positive changes also made.