PhonoCommand inference reference, Apache-2.0.
qwen_asr_backend/* and asr_output.py derive from Qwen3-ASR SDK 0.0.6, Alibaba Qwen team, Apache-2.0.
https://github.com/QwenLM/Qwen3-ASR
Official model/config/processor files retain their copyright headers.
asr_output.py extracts unchanged parser/repetition functions; chunk/prompt adaptation is in asr_windows_native.py.