kokoro-tts

Converts text to speech using local models on Apple Silicon.

2|Updated Apr 6, 2026
One-click install
npx skills add https://github.com/bolander72/bstack --skill kokoro-tts-bolander72
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kokoro-tts
Source: https://github.com/bolander72/bstack/tree/main/skills/kokoro-tts
Command: npx skills add https://github.com/bolander72/bstack --skill kokoro-tts-bolander72

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Generate speech audio from text using local models on Apple Silicon. It runs offline with zero API keys and no internet access.

Core Features & Use Cases

  • Local Text-to-Speech with two models: pocket-tts-4bit (mlx-audio) and kokoro-onnx (Kokoro-ONNX).
  • Supports voice selection, speed control, and language options for applications like voice messages, in-app TTS, and accessibility.
  • Fully offline operation with fast turnaround on supported hardware.

Quick Start

Generate speech from text using your preferred local model (mlx-audio pocket-tts-4bit or Kokoro-ONNX) and save the output as a WAV file.

Frequently Asked Questions about kokoro-tts

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate natural text-to-speech audio locally on macOS without an internet connection?

Local text-to-speech on macOS can be generated offline using models like mlx-audio pocket-tts-4bit or Kokoro-ONNX, converting text into natural speech audio files without requiring internet access or API keys.

Can I use offline TTS on Apple Silicon to create voice messages with custom voices?

Offline TTS on Apple Silicon supports voice selection and speed control, allowing you to create customized voice messages and in-app audio responses entirely locally.

Does local text-to-speech on macOS require any API keys or internet access to function?

Local text-to-speech using Kokoro or mlx-audio models requires zero API keys and no internet access, running entirely offline on supported Apple Silicon hardware.

What is the best way to convert text into WAV speech files locally using Kokoro?

The best way to convert text into WAV speech files locally is by using the Kokoro-ONNX or mlx-audio pocket-tts-4bit models, which support offline voice generation and speed control.

Are there language options available when generating offline speech with mlx-audio on macOS?

Offline speech generation with mlx-audio includes optional language settings alongside voice selection and speed control, supporting diverse in-app TTS and accessibility use cases.