Swarit models
A voice for
every conversation.
Open-source models to explore. Platform models to build with. Speech technology for Indian languages and English.
Open source · Published
Swarit Indic Talk.
Our voice-to-voice model for Indian languages and English.
Explore the published model on Hugging Face for model files, license information, and release details.
View the model on Hugging Face →Platform models · Realtime
Realtime conversation model.
Configure instructions, language, voice, reflex answers, and conversation steps through the developer API. Realtime conversations are priced at ₹2 per minute.
See the integration preview →Platform models · Coming soon
Text-to-speech.
Turn text into spoken audio. Our standalone TTS offering will let developers add speech generation to their applications, independently of a realtime conversation.
Streaming API access, model details, and pricing are coming soon. We’re updating the infrastructure and architecture before opening access.
Platform models · Coming soon
Speech-to-text.
Turn spoken audio into text. Our standalone STT offering will let developers add speech recognition to their applications, independently of a realtime conversation.
Streaming transcription, model details, and pricing are coming soon as we prepare the standalone service.