Puzzle Multilingual ASR Test

Model: qwen3asr_small_v22_9ep_7292 /model
Task
Languages (none = auto-detect)
Context / Prompt
off
off
On silence after speech, the server finalizes the segment (commits the partial) automatically.
off
Restrict output to the selected language's script (needs exactly one language). E.g. English then never emits Korean/Arabic/Japanese.
Input Gain (applies to mic & file, client-side)
0 dB
Audio File (decoded → 16 kHz mono, gain applied, streamed through model)
Idle
-∞ dBFS Clear history
Detected Language
Live Transcription
History