Generates offline text-to-speech audio responses using local Piper voice models and sherpa-onnx.
This skill enables Claude to respond with natural-sounding spoken audio messages by leveraging the Piper TTS engine and sherpa-onnx runtime. Designed for privacy and performance, it processes text-to-speech entirely on your local infrastructure, requiring no external API keys or internet connection once setup. It features multi-language support for English and German, automatic language detection, and outputs specialized tags for Telegram compatibility, allowing audio files to appear as native voice notes in messaging interfaces.
主な機能
01Automatic language detection for seamless multi-lingual replies
023 GitHub stars
03100% offline local TTS processing for maximum privacy
04Zero-cost operation with no API keys or subscriptions required
05Support for high-quality natural male voices in English and German
06Telegram-ready output formatting for native voice bubble display
ユースケース
01Generating spoken responses for Telegram-based AI assistants
02Creating audio versions of documentation or text for accessibility
03Building offline-capable voice interfaces for local development workflows
