Lightning Whisper MLX

A batched, quantized Whisper reimplementation chasing raw transcription speed.

When to use it

Use this when throughput on long audio matters more than having an actively maintained dependency.

Text-to-speech, speech-to-text and audio models under one MLX package.

Prompt-controllable speech: describe the voice you want in plain English.

Speech in, speech out, skipping the ASR-LLM-TTS relay race.