The voice interface for the next 5 billion Backed By @southpkcommons
Joined May 2025
- Tweets45
- Following9
- Followers4.8K
- Likes492
Pinned Tweet
Understanding how india 🇮🇳 actually speaks has been one of the hardest problems in technology.
1.4 billion people. the most linguistically diverse and complex country on earth.
today, we’re showing maya understand the way india actually speaks.
Maya Research retweeted
Congratulations to @Dheemanthredy and the @mayaresearch_ai team on releasing Maya-2-Native's weights under Apache 2.0: streaming TTS for 11 Indic languages and English.
On the Voice Arena Hindi TTS Leaderboard it ranks #2 of 9 among real-time models and #7 of 20 overall.
Maya-2 voice model weights are officially live! 🇮🇳🇮🇳
The frontier voice model for India is now completely open-weight.
• Outperforms ElevenLabs v3, Sarvam Bulbul v3, and Cartesia Sonic 3.5
• 1.5B+ people covered across all 28 states and 8 union territories
• SOTA voice synthesis across 11 official languages
• 20 native speakers pre-loaded, ready out of the box
• Best-in-class zero-shot voice cloning
Built by a team of five from the land of languages.
Go build and have fun!
Maya Research retweeted
Maya-2 voice model weights are officially live! 🇮🇳🇮🇳
The frontier voice model for India is now completely open-weight.
• Outperforms ElevenLabs v3, Sarvam Bulbul v3, and Cartesia Sonic 3.5
• 1.5B+ people covered across all 28 states and 8 union territories
• SOTA voice synthesis across 11 official languages
• 20 native speakers pre-loaded, ready out of the box
• Best-in-class zero-shot voice cloning
Built by a team of five from the land of languages.
Go build and have fun!
👨🍳👨🍳👨🍳
We’re releasing the world’s best Indic voice AI model, fully open-weight. 🇮🇳🇮🇳
With over 1.5 billion people and 22 official languages, the future of AI for India is voice-first.
Maya-2 officially outperforms ElevenLabs v3, Cartesia Sonic 3.5, and Bulbul v3 across Indian languages.
It will be completely free for every builder very soon.
Build with maya voice
Too much infighting on languages, so I vibe coded a web app that lets you click a picture of an object & it ll teach you how to pronounce it in Tamil, Gujarati, Kannada, Malayalam & Marathi. Took less than 2 hrs.
Built w @openai & @mayaresearch_ai
see-it-say-it.vercel.app/
Maya Research retweeted
Amazing news on India's Independence Day 🇮🇳
Maya 2 Native is ranked #1 for Hindi text-to-speech on Voice of India, the human-evaluated voice arena from @ai4bharat, @iitmadras and @JoshTalksLive
The only Made-in-India model at the top. Ahead of Google's Gemini 2.5 Pro and Cartesia's Sonic-3.5, decided by 21,770 real human votes.
We started Maya at @spc in Bengaluru with one belief: the future of everything is voice, and India, the land of languages, will be the leader in building it for the world.
Today our own model is #1 in our own language, at home. Our team has trained models that hold their own against the hardest leaderboards in the world, and we are just getting started.
This is the beginning of something meant to outlast us.
voiceofindia.ai/leaderboards…
You are the best judge of a voice model’s quality.
Voice is human. The leap between today’s voice models and Maya Research’s new models is something no leaderboard can capture. You have to hear it for yourself.
Maya Research retweeted
Maya-2-Native is the #1 streaming TTS model on the Voice Arena Hindi TTS Leaderboard 🇮🇳
In the overall leaderboard (streaming + non-streaming), a native Indian voice lab has officially broken into the top tier, debuting at rank #2 globally, moving ahead of global heavyweights like ElevenLabs v3 and Cartesia Sonic-3.5.
Maya-2-Native is the latest TTS model from @mayaresearch_ai. Built locally to capture the true texture of everyday Indian speech, it handles complex code-switching, local dialects, and natural cadence effortlessly. The model has been highly preferred among native speakers rating on Voice Arena.
Results backed by 15.5K+ human-verified, blind, head-to-head listener votes 🧵
Maya Research retweeted
.@Dheemanthredy says the future of AI won't be won by the first 50 million users. It'll be won by the next 5 billion.
"We're not building Maya for the top-tier consumers. We're building it for the five billion people."
"They haven't experienced technology the way we have."
"Imagine putting that technology in your grandmother's hands, and she's talking to it, doing everything you're doing."
"I'm so scared of not living in that future."
@mayaresearch_ai
Maya Research retweeted
Excited to share that Maya Research has raised $1.9M from South Park Commons, our first partners in this journey.
Maya is building the voice interface for the rest of the world. For the billions who live through voice.
We are building this interface around how people actually talk in real life, how they have conversations, discover, and decide who to trust.
The rest of the world will only use the technology at this scale when it feels native, like one of their own.
@spc @spc_india
🤖 Made with AI
maya now understands the way we speak: अपनी हिंदी, మన తెలుగు, நம்ம தமிழ், ನಮ್ಮ ಕನ್ನಡ, നമ്മുടെ മലയാളം, আমাদের বাংলা, आपली मराठी, ਸਾਡੀ ਪੰਜਾਬੀ, આપણી ગુજરાતી, আমাৰ অসমীয়া, and more.
try it here:
mayaresearch.ai/blog/maya-li…
voice interface is how billions of people will access ai.
the core problem is conversational ability. models that can hold real conversations in accents and dialects, the way people actually speak in their daily lives.
listen to maya’s latest voice interface.
Maya Research retweeted
Inside Maya Research: How two 23-year-olds beat Google in speech AI and are building a 10 to 15x larger Indic version by June 2026
@Dheemanthredy @bharathkmni @deedydas @spc
moneycontrol.com/artificial-…