Transcribe audio files locally using MLX on Apple Silicon.
npx skills add https://github.com/guoqiao/skills --skill mlx-stt
Please help me install this Agent Skill. Skill: mlx-stt Source: https://github.com/guoqiao/skills/tree/main/mlx-stt/mlx-stt Command: npx skills add https://github.com/guoqiao/skills --skill mlx-stt
This Skill provides a local, private, and free Speech-to-Text (STT) and Automatic Speech Recognition (ASR) solution, eliminating the need for external APIs or servers.
Transcribe the audio file located at '/path/to/your/audio.mp3' into text.
High-intent search queries and answers about installing and using this skill.
Local speech-to-text on Apple Silicon transcribes audio files directly on your machine using MLX, ensuring your data remains completely private without requiring external APIs or internet access.
Running local speech-to-text with MLX requires ffmpeg for audio conversion, uv for execution, and mlx-audio for model operations on Apple Silicon to function properly.
Offline speech-to-text works for transcribing recorded meetings on Apple Silicon Macs, processing audio files locally after the initial model download without needing an active internet connection.
Local ASR with MLX supports offline transcription completely, requiring an internet connection only for the initial model download before processing audio files entirely on device.
The best way to perform private speech-to-text on macOS is using MLX with open-source models, which processes audio locally on Apple Silicon hardware without uploading data to external servers.
guoqiaoGenerate speech audio from text locally on macOS Apple Silicon using MLX models.
sahiixxGenerate embeddings, analyze images, and extract structured JSON from text and documents.
rustykuntzTranscribe audio to TXT, VTT, or SRT locally with Parakeet MLX.
surus-latBenchmark Whisper-family ASR models on FLEURS datasets with WER and CER metrics.
