Maya-2 debuts as open-weight Indic voice model, ranks #2 on Hindi TTS leaderboard
CurieuxExplorer · x · 2026-10-15
Maya Research released Maya-2, billed as the world's best open-weight Indic voice AI model.
- The team says Maya-2 outperforms ElevenLabs v3, Cartesia Sonic 3.5 and Sarvam Bulbul v3 across Indian languages, and will be free for all builders soon
- On the VoiceArena Hindi real-time TTS leaderboard, Maya-2-Native sits at #2 with 1049 Elo, behind only Cartesia Sonic-3.6 (1097) and well above Bulbul-V3 (#7, 959) and Grok TTS (#8, 893)
- Rationale: with 1.5B+ people and 22 official languages, the team argues India's AI future is voice-first, targeting education, healthcare and regional-language digital services
Related event: Maya-2 Open-Weights Speech Model Claims Top Spot for Indic TTS(2 posts)→
More from Models
- UltraBERT: open 150M/380M encoders with 32K context and an unusual dual-masking recipe — algo_diver · 2026-10-15
- DeepSeek V4.1 measures its own decode speed at 430-460 tok/s — then the claim falls apart — teortaxesTex · 2026-10-15
- OpenCode offers $60 of Gemini 4 Argon usage for $10 until Nov 30 — m4rkmc · 2026-10-15
- After losing 5 Claude accounts, one user's controlled tests show what actually triggers bans — jiayuan_jy · 2026-10-15
- Dev Says Jev Could Easily Detect X Tweet Spam, But X Hasn't Prioritized It — ivan_bezdomny · 2026-10-15
- DeepSeek outage ends, sparking new-model rumors that the poster himself retracts as false alarm — teortaxesTex · 2026-10-15