# Real-time Speech Processing

Lite-Whisper is a compressed version of OpenAI Whisper, utilizing LiteASR technology to reduce model size while maintaining high accuracy.

Ultravox is a multimodal voice large language model based on Llama3.2-1B and Whisper-large-v3, capable of processing both voice and text inputs.

A speaker diarization model based on pyannote.audio 2.0, designed for automatic detection and segmentation of different speakers in audio.

This is a speech enhancement model trained using the MetricGAN+ method, capable of effectively improving speech quality.