Expressive Indian-language speech, built for real time
Bulbul V4 Flash is Sarvam AI's text-to-speech model built for natural, expressive speech across Indian languages. It delivers under 250 ms time-to-first-audio, with 83 voices across 11 languages and 224 voice-and-style combinations. It's up to 40% faster than Bulbul V3 under comparable load and costs ₹20 per 10,000 characters. Try it through the Sarvam API, Playground, or Content Studio.
Bulbul V4 Flash is a text-to-speech model by Sarvam AI that offers expressive speech in 11 Indian languages with 83 voices and 224 voice-and-style combinations. It features a time-to-first-audio of under 250 ms and is 40% faster than its predecessor, Bulbul V3, at a cost of ₹20 per 10,000 characters.
A source that found nothing is a measurement. A source that has not run is a gap. Neither means the launch lacks the thing.