Saltar al contenido
TakoAPI
Volver a skills

local-voice

por trondw

DevOps & Cloud
1.2kDescargas según clawskills.sh
1Estrellas
0Vistas
0 me gustaInicia sesión para dar me gusta a esta skill
—Aún sin valoraciones
Tu valoración

Resumen

Local text-to-speech (TTS) and speech-to-text (STT) using FluidAudio on Apple Silicon.

Descripción

Qué hace esta skill

Runs text-to-speech and speech-to-text entirely on Apple Silicon using FluidAudio's CoreML models via the Apple Neural Engine. TTS uses the Kokoro model with 54 voices at ~0.6-0.8s latency; STT uses Parakeet TDT v3 with ~0.2-0.3s latency across 25 languages. No network connection or API key required.

Achieves sub-second voice synthesis and transcription on-device, replacing cloud TTS/STT services without API costs, rate limits, or sending audio data off-device.

Cuándo usarla

  • Reading long articles aloud while multitasking
  • Transcribing recorded meetings without uploading audio to the cloud
  • Adding offline voice output to a local AI assistant
  • Converting written notes to audio files for listening on the go
  • Building voice interfaces for local automation scripts

Flujo de trabajo de ejemplo

ENTRADA

User asks: Read this paragraph aloud

AGENTE
  1. 1

    Receives text input from the user

  2. 2

    POSTs text to the local TTS endpoint at http://127.0.0.1:18790/synthesize

  3. 3

    StellaVoice daemon processes the request using the Kokoro model on the Apple Neural Engine

  4. 4

    Returns a WAV file in approximately 0.6-0.8 seconds

  5. 5

    Plays the audio output to the user

SALIDA

The paragraph is spoken aloud in the selected Kokoro voice with natural pacing

Requisitos

  • Apple Silicon Mac (M1 or newer) running macOS 14+
  • espeak-ng (install via Homebrew: brew install espeak-ng)

Instalación

clawhub install trondw/local-voice

Ejecuta el comando de la CLI o pega el prompt en Claude Code, Cursor, Codex o el agente de programación que uses.

Enlaces