Maya (Maya Research)
Open-source emotionally intelligent voice AI built for the 90% of the world mainstream TTS ignores.
Overview
Maya-1 is a 3B-parameter open-source TTS model from Bengaluru's Maya Research, built on a Llama-style transformer with 20+ emotion tags and natural-language voice design. Released Apache 2.0 in Nov 2025, it runs on a single 16GB GPU and ranks near #6 among open-weight models on Artificial Analysis.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 1 Sep 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- Full model access
- Commercial use allowed
- No vendor lock-in
- No self-hosting needed
- Playground + docs
Use cases
What you can produce with Maya (Maya Research)
- Emotionally expressive TTS audio at 24 kHz with sub-100 ms latency
- 20-plus emotion-tagged voice outputs (laugh, cry, whisper, angry, gasp, giggle, sigh, and more)
- Zero-shot voice generation from natural-language persona descriptions (age, accent, tone, character)
- Self-hosted inference stack compatible with vLLM, llama.cpp GGUF, and ComfyUI nodes
- Open-weight Apache 2.0 model weights downloadable from Hugging Face (maya-research/maya1)
- Hosted web studio demo at mayaresearch.ai/studio for no-code voice previewing
- API access via console.mayaresearch.ai for production integration
ASEAN Perspective
Maya (Maya Research) in Southeast Asia
Maya Research explicitly targets Southeast Asia as a key market alongside India and MENA, and the consumer app has already crossed 3 million installs across these three regions combined. The company's mission — building voice AI for the 90% of the world that mainstream English-first systems ignore — maps directly onto ASEAN's linguistic diversity, where hundreds of regional languages and accents are underserved. Current open-source weights cover English; Hindi and Telugu are the next confirmed languages, with broader Indic and Asian language support planned for mid-2026. ASEAN developers can self-host Maya-1 today for English-language voice apps at zero per-call cost, making it a strong candidate for regional accessibility and low-bandwidth use cases.
Maya-1 is a genuinely impressive technical feat from a two-person Indian startup that trained a globally competitive open-weight TTS model on free cloud credits alone. At launch (November 2025) it ranked #2 among open-weight models on the Artificial Analysis Speech Arena; the field has crowded in since, and by mid-2026 it sits closer to #6 — its 1,051 Quality Elo hasn't moved, but rivals have caught up. The 20+ emotion tags and zero-shot natural-language voice design still stand out. For self-hosters, Apache 2.0 means no per-character fees and no vendor lock-in, which matters in cost-sensitive ASEAN deployments. Caveats: the open weights cover English with multi-accent support only — Hindi and Telugu are still in development — self-hosting needs a 16GB-VRAM GPU, and hosted API pricing at console.mayaresearch.ai isn't publicly documented.
What people say
Two 23-year-old NYU graduates from rural Andhra Pradesh built a text-to-speech model on free cloud credits that briefly out-ranked entries from labs with nine-figure budgets. Maya-1, released under Apache 2.0 in November 2025, hit #2 among open-weight models on the Artificial Analysis Speech Arena at launch — a genuine result, not a cherry-picked benchmark.
That rank hasn't held: by mid-2026 Maya-1 sits closer to #6 on the same leaderboard, even though its own Quality Elo score (1,051) hasn't changed. The field simply got more crowded, which is a compliment to the category Maya helped kick off as much as a knock on the model. What hasn't changed is the pitch — 20-plus inline emotion tags (whisper, laugh, gasp, cry), zero-shot voice design from a plain-language description like "warm 40-year-old male, slight Southern Indian accent," and sub-100ms latency on a single 16GB GPU.
Traction is real for a company this young: 440,000+ Hugging Face downloads and a companion consumer app past 3 million installs across India, Southeast Asia, and MENA, on a $1.9M seed round led by South Park Commons. The open-weight release is English-only with multi-accent support; Hindi and Telugu are still in development, and the hosted API at console.mayaresearch.ai has no public pricing page. Self-hosting needs a 16GB-VRAM GPU, which rules out casual tinkering on a laptop.
Best read as a genuinely strong open-weight option for cost-sensitive ASEAN builders today, with the multilingual roadmap the thing to watch through the rest of 2026.
Summary of public user & expert reviews, compiled by RECATOOLS.
Notable facts
- Maya Research built the entire Maya-1 model without venture capital, using only free cloud credits from AWS, Google, and Microsoft Azure.
- Both co-founders are 23-year-old NYU graduates from rural Andhra Pradesh — Dheemanth Reddy from Vempalli and Bharath Kumar Kakumani from Ongole.
- Maya Research is the only Indian company listed on the Artificial Analysis Speech Arena leaderboard as of 2025.
- Maya-1 supports over 20 inline emotion tags that physically alter the audio waveform — changing pitch, breath timing, and pacing — rather than just applying a filter on top.
Frequently asked questions
About this listing
This entry was compiled from publicly available data including Maya (Maya Research)'s official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Maya (Maya Research) unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Maya (Maya Research) directly →
Spotted something out of date? Suggest an update →
Alternatives to Maya (Maya Research)
More in LLMs & Chat
