Maya Research Releases Open-Weight Speech Model Supporting 12 Languages
Maya 2 Native is an Apache 2.0-licensed open-weight text-to-speech model that generates speech in Assamese, Bengali, English, Gujarati, Hindi, Kannada, Malayalam, Marathi, Odia, Punjabi, Tamil and Telugu. The model streams 24 kHz mono audio in 80-millisecond frames for real-time playback, includes 20 built-in synthetic voices and supports voice cloning using short audio samples. Developers can use the model to build multilingual voice assistants and regional-language AI services, with cloned voices able to generate speech in different languages than the original recording.
Key points
- Open-source TTS model for 11 Indian languages plus English with real-time streaming
- Supports voice cloning with cross-language generation capabilities
- Includes 20 synthetic voices; designed for voice assistants and regional apps
Summary and key points written by AI from the headline and a short excerpt, so they can be incomplete or wrong. Analytics India Magazine did the reporting; read the original for the full story.
Read the full story at Analytics India Magazine
