Generates high-quality speech from text locally using Sherpa-ONNX models without requiring cloud connectivity.
The sherpa-onnx-tts skill provides a robust, privacy-first solution for text-to-speech directly within your AI assistant environment. By leveraging ONNX-based models like Piper, it enables users to generate audio files from text strings entirely offline. This is ideal for developers and users requiring voice synthesis while maintaining data sovereignty, working in air-gapped environments, or seeking to avoid the recurring costs and latency associated with cloud-based TTS APIs.
Key Features
01Customizable model paths for easy switching between different voices
02Cross-platform compatibility for macOS, Linux, and Windows
03Support for high-quality VITS-based models such as Piper
04Command-line interface for direct audio generation to WAV files
058,363 GitHub stars
06100% local execution for maximum privacy and offline accessibility
Use Cases
01Integrating local speech synthesis into developer notification systems
02Automating voiceover generation for documentation or tutorials
03Generating audio content in privacy-sensitive or low-connectivity environments
