AI Model Inference Platform

Production-ready ML models for real-time inference and experimentation

6 Models Active

Average latency: 193ms

Select Model

Speech-to-Text for Khmer language with dialect support

Speech RecognitionLatency: ~89msEndpoint: /api/inference/asr-stt

Model Input

Ready

Upload audio file

WAV, MP3, M4A (max 10MB)

Enable Punctuation

Enable Speaker Diarization

Inference Output

Output will appear here after running inference

Model Information

Statusonline

Input Typeaudio

Output Typetext

Avg Latency89ms