Pocket TTS
Generates high-quality local text-to-speech audio directly from Claude Code using a low-latency streaming server.
Pocket TTS enables Claude to communicate audibly by leveraging a local text-to-speech server, eliminating the need for external cloud APIs. It supports real-time streaming playback for immediate audio generation, a variety of built-in voice profiles, and the ability to create custom voice embeddings from short audio clips. Designed for macOS, this skill is ideal for users who want to hear code explanations, documentation, or narratives read aloud while multitasking, providing a seamless auditory layer to the AI coding experience.
主な機能
014 GitHub stars
02Built-in library of diverse voice profiles including 'Alba' and 'Javert'
03Seamless integration with ffplay and macOS afplay utility
04Local low-latency audio streaming for immediate playback
05Custom voice cloning support via 30-second audio samples
06Persistent server architecture to keep models warm in memory
ユースケース
01Creating narration for scripts or stories generated within the terminal
02Accessibility enhancement for users who prefer auditory feedback over reading
03Audible code reviews and documentation narration while multitasking