ailey-tools-audio

You can extract audio from video and transcribe it by running a multi-step pipeline that demuxes the audio track and sends it directly to Whisper for transcription.

You can convert across over 10 audio formats including mp3, wav, flac, aac, and ogg using FFmpeg, ensuring compatibility with downstream transcription services.

Silence-based slicing optimizes Whisper transcription by splitting audio files at silent points, allowing batch processing of smaller segments to improve accuracy and efficiency.

Yes, you can transcribe multiple audio files in batch using the CLI, which applies input validation and error handling to process single or multiple files with sensible defaults.

Yes, you need an OpenAI API key configured in your environment to use Whisper for transcription, alongside FFmpeg installed for audio processing and silence detection.

Audio demux pipelines fail during format conversion due to invalid input formats or missing dependencies, but the CLI provides robust error handling to identify and report these issues.