Please consider Kyutai Pocket-TTS- it's an efficient 100M model with good quality, and has a voice cloning feature. https://kyutai.org/blog/2026-01-13-pocket-tts (samples) https://github.com/kyutai-labs/pocket-tts
Please consider Kyutai Pocket-TTS- it's an efficient 100M model with good quality, and has a voice cloning feature. https://kyutai.org/blog/2026-01-13-pocket-tts (samples) https://github.com/kyutai-labs/pocket-tts