Audio / Text-to-Speech

Bulbul V4 Flash

Text-to-speech model from Sarvam AI · released 2026-10-13 · vendor-reported (not yet ranked on an independent leaderboard)

Unrankednot on an independent leaderboard yet
2026-10-13release

Specifications

Languages
11 (10 Indian languages + English)
Voice catalogue
200+ persona voices (222 voice IDs) in Sarvam's docs; launch coverage cited 83 voices
Input limit
2,500 characters per REST request
Launch pricing (reported)
₹20 per 10,000 characters
Latency (vendor claim)
under 250 ms time-to-first-audio
API model id
bulbul:v4-flash

Notes

  • Not ranked on the Artificial Analysis Speech Arena at the time of writing; all performance claims are Sarvam's own.

How this is measured

Speech Arena (Text-to-Speech)

Elo rating (higher is better)

Models are ranked with an Elo system derived from blind user votes: listeners hear two speech samples generated from the same text and pick the one that sounds more natural. The provider-voice leaderboard uses each provider's own native voices. Each result is reported with a 95% interval and the number of votes (samples).

Release & sources

Release
2026-10-13
Release source
Launch coverage (Analytics Insight, 13 Oct 2026); Sarvam docs list it as the latest TTS model

More TTS models from Sarvam AI