Transcribe speech and synthesize text with Rust and candle.
npx skills add https://github.com/rgbkrk/voice --skill voice-rgbkrk
Please help me install this Agent Skill. Skill: voice Source: https://github.com/rgbkrk/voice/tree/main Command: npx skills add https://github.com/rgbkrk/voice --skill voice-rgbkrk
This Skill enables natural language interaction with text, provides real-time speech-to-text transcription, and offers efficient ways to manage audio data.
Start the voice daemon with voice daemon start and use the following commands:
voice say "Hello world"voice listenvoice transcribe recording.wavHigh-intent search queries and answers about installing and using this skill.
You can transcribe audio files to text using AI by running the command `voice transcribe recording.wav`. This Speech-to-Text functionality processes audio input and outputs the corresponding text transcription.
Setting up real-time speech-to-text and TTS requires installing Rust and associated libraries, along with the candle dependency. The environment supports speech synthesis on Apple Silicon and provides a CPU fallback on Linux.
To make your text application speak output aloud, start the voice daemon and use the `voice say` command. This Text-to-Speech feature supports customizable voices and speed for spoken audio output.
Yes, the TTS and STT solution works on Linux by utilizing a CPU fallback for speech processing. It also offers optimized speech synthesis support for Apple Silicon architectures.
You can use natural language voice commands to control transcription and speech synthesis. By starting the daemon, the `voice listen` command enables interactive AI systems to capture spoken input and execute voice commands.
simorgh3196Enforce domain-level constraints for Rust CLI tools using clap.
TIAN0517Transcribe spoken audio from files or base64 inputs into text.
TikTzukiIntegrate LLMs, STT, and TTS services with WebSocket audio streaming.
frankhli843Synthesize speech audio locally from text using sherpa-onnx.