• Initial release of Pocket TTS Skill.
• Provides fully local, offline text-to-speech using Kyutai’s Pocket TTS model—no API calls required.
• Supports 8 built-in voices plus custom voice cloning from WAV files.
• Features fast CPU-only performance (~6x real-time) and low-latency streaming output.
• Includes CLI and Python API for flexible integration.
• Requires accepting the model license on Hugging Face before installation.