Sarvam AI Launches Bulbul-v2 Text-to-Speech Model for Indian Languages
| Aspect | Details |
|---|---|
| Event | Bengaluru-based AI startup Sarvam AI launched Bulbul-v2, a text-to-speech (TTS) model. |
| Key Feature | Supports 11 Indian languages with regional accent precision. |
| Technical Features | - Real-time synthesis and multi-language (including code-mixed) text support.<br>- Fine-grained control over pitch, pace, loudness, and sample rates (8kHz to 24kHz).<br>- Smart text preprocessing: normalises numbers, dates, and mixed-language content. |
| Aim | - Democratise AI voice technology for Indian users.<br>- Promote linguistic inclusivity in India's digital ecosystem. |
| Background | - Bulbul-v1 launched in August 2024 with six preset voice personalities.<br>- Sarvam AI is developing India's sovereign large language model under the IndiaAI mission. |
| Significance | - Improves accessibility for digital services in local languages.<br>- Enables brands to reach regional audiences with authentic-sounding voices.<br>- Boosts India's AI ecosystem, supporting technological self-reliance. |