KaniTTS2-en

KaniTTS2-en is an open-source text-to-speech model developed by NineNineSix, optimized for real-time conversational AI applications. It utilizes a two-stage pipeline combining a large language model with an efficient audio codec to deliver high-quality, low-latency speech synthesis.

KaniTTS2-en is an open-source text-to-speech model developed by NineNineSix, optimized for real-time conversational AI applications. It utilizes a two-stage pipeline combining a large language model with an efficient audio codec to deliver high-quality, low-latency speech synthesis.

• 01Two-stage pipeline combining LLM and FSQ audio codec
• 02Model size of 400 million parameters
• 03Supports English language with a sample rate of 22kHz