🎙️ Qwen3-ASR — speech-to-text

⚠️ Public demo — do not upload confidential documents or audio.

Speech recognition for 52 languages and dialects with Qwen3-ASR (Apache-2.0). Running on ZeroGPU.

Audio (upload or record)

Model

Language

Also produce .srt timestamps (forced aligner; en/zh/yue/fr/de/it/ja/ko/pt/ru/es only, ≤3 min)

Transcript

Download .txt

Download .srt (if timestamps)

⚠️ Public demo — do not upload confidential documents or audio.