Voice TeamSeptember 09, 2026•3 min read
A New Chapter for the Miransas Voice Team
Our voice program continues with Guliruhsar now leading recordings, with Malika set to join the team soon.
Read full announcement
Miransas Newsroom & Research
Voice AI Research, Releases & Insights
Follow our journey building real-time multilingual voice models, native Uzbek dataset engineering, and next-generation speech pipelines.
Featured Update
Latest Research & Updates
Research5 min read
Why Uzbek Is a First-Class Training Track for Miransas
Instead of treating Uzbek as a translation afterthought, Miransas is building language-specific data, phoneme coverage and evaluation around native speech.
August 31, 2026
Multilingual4 min read
Adding More Global Languages to the Evaluation Lab
English, Spanish, Chinese, Hindi, Arabic, Japanese, Korean, French, German, Portuguese, Turkish and Russian are now part of the broader comparison set.
August 28, 2026
Models6 min read
Miransas and the Next Generation of Voice AI
We are comparing Miransas with GPT-Realtime, Gemini Live, Grok Voice and the Chatterbox baseline using transparent capability categories rather than invented leaderboard numbers.
August 25, 2026
Engineering7 min read
Inside the Miransas Training Pipeline
From clean speech data and speaker embeddings to evaluation and inference, this is the direction behind our next voice models.
August 22, 2026
Model Capabilities & Baseline Stats
Miransas baseline
500M
Chatterbox Multilingual V3
Baseline coverage
23+
Chatterbox multilingual languages
Audio
Realtime
Voice AI evaluation stack
Uzbek
Native
Miransas training direction
Transparent Capability Matrix
Direct comparison of foundational features without inflated benchmarks.
| Capability | Miransas | GPT-Realtime | Gemini Live | Grok Voice |
|---|---|---|---|---|
| Realtime audio | ||||
| Audio input / output | ||||
| Voice cloning | — | — | — | |
| Open-source baseline | — | — | — | |
| Custom language training | — | — | — | |
| Uzbek training track | — | — | — |
Languages & Uzbek Training Track
Native Speech Focus
Uzbek as a First-Class Language Track
Most global voice models handle regional languages through generic translation layers. Miransas builds native phoneme coverage, local dialect accents, and custom dataset pipelines specifically optimized for Uzbek alongside 23+ global languages.
Uzbek (Native Track)EnglishTurkishSpanishChineseArabicHindiGermanRussian+ 15 More