Kyutai has released three smaller versions of their Pocket TTS models, with parameter counts of 1.9M, 6.9M, and 18.2M. These English-language models are capable of voice cloning and are designed to run on a central processing unit (CPU). This allows users to keep speech generation on their CPU, freeing up GPU resources for large language models. AI
IMPACT Enables on-device voice generation, freeing up GPU for LLMs in local assistant setups.
RANK_REASON Release of smaller, CPU-runnable TTS models for voice cloning.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →