Interactive Sandboxes

See it. Hear it.

Hands-on demos of Tynrai's language models — translate text, synthesize natural speech, and transcribe audio for the languages of Meghalaya.

Translation

Neural Machine Translation

  • English ↔ Khasi bidirectional NMT
  • English → Pnar & English → Garo
  • Voice dictation, TTS playback & document OCR
Launch NMT Sandbox

Text to Speech

Give your words a local voice

  • Khasi, Garo, Pnar, Hindi & English
  • Natural male & female voice profiles
  • Fine-grained prosody & guidance controls
Launch Rynsan TTS

Speech to Text

Turn spoken audio into text

  • Khasi acoustic transcription
  • Upload audio or record live in browser
  • Timestamps & multi-speaker diarization
Launch Rynsan ASR

Models run on Tynrai's own Hugging Face Spaces. Explore the model weights →