Skip to content

Open model

Hojo ASR

by HojoAI

Open-weight conversational speech recognition models from HojoAI that decode with a Qwen3 language model and handle Chinese-English code-switching, run with the hojo-asr Python package.

  • Self-hosted
Official siteDocumentation

Every figure about Hojo ASR

Each figure is a row of a research dataset. Open it for the exact words of its source or the exact request it was read from.

Speech Recognition Accuracy: Word Error Rates

Speech Recognition Speed: Real-Time Factor Data

Open ASR Leaderboard: Speech Recognition Accuracy, Tracked

Other open models and engines

All open models and engines, compared