Coqui

Coqui is a voice AI platform offering advanced text-to-speech and voice cloning tools. Its core technology includes the ⓍTTS model, which enables voice cloning from just a few seconds of audio, supports cross-language voice reproduction, and offers streaming inference with very low latency. It’s designed for developers, researchers, and creators who need high-quality synthetic voices with flexibility and control. 

Key Features:

• Voice cloning from extremely short audio samples (as low as ~3–6 seconds).

• Multilingual speech generation and cross-lingual voice cloning.

• Streaming inference with latency under 200 ms and high sampling rate (24 kHz)

• Open-source toolkit with support for training/fine-tuning custom models and wide deployment options