Live Demo
Turn speech into text.
Upload or record Khasi audio and have it transcribed.
Preview mode. The interface is running against a local stub — no model is being called yet.
Audio
Transcription currently supports Khasi audio only.
0:00
Tap to record · stops automatically at 30 seconds
Output format
Advanced parameters
Skips silence so the model doesn't guess words there.
Higher = fewer repeated words, but may skip some entirely.
Higher = more accurate search, but slower to transcribe.
How long a pause must be before it's treated as a break.
How sure the model must be that a part is silent before skipping it.
Appended to the prompt to bias recognition toward these terms.
Transcript
The transcript will appear here.
Models run on Tynrai's own Hugging Face Spaces. See the models