#7901 No audio file was generated during the listening test.self=[QwenttsLocal]: uuid=None, proxy_str=None, tts_type=1, len=1,

37.142**1 Posted at: 2 hours ago

No audio file was generated during the listening test.
self=[QwenttsLocal]: uuid=None, proxy_str=None, tts_type=1, len=1, language='en', play=True, volume='+0%', pitch='+0Hz', rate='+0%', speed=1.0, has_done=0, wait_sec=1.0, dub_nums=1, error='No audio file was generated during the listening test.', api_url='', is_cuda=False, local_dir='D:/AI_dub_pyvideotrans-v4.15/models/models--Qwen--Qwen3-TTS-12Hz-0.6B-CustomVoice', is_redubb=False, model_name='0.6B', ainame=None:
{'text': 'Hello, my dear friend, I hope every day of yours is wonderful and joyful!', 'rate': '+0%', 'role': 'zh_male_bj.wav', 'filename': 'D:/AI_dub_pyvideotrans-v4.15/tmp/54532/listen_voice/1-en-1791526474.675175.wav', 'tts_type': 1, 'language': 'en', 'volume': '+0%', 'pitch': '+0Hz'}
Traceback (most recent call last):
File "videotrans\util\ListenVoice.py", line 21, in run
File "videotrans\tts\__init__.py", line 122, in run
File "videotrans\tts\_base.py", line 125, in run
videotrans.configure.excepts.DubbingSrtError: No audio file was generated during the listening test.
self=[QwenttsLocal]: uuid=None, proxy_str=None, tts_type=1, len=1, language='en', play=True, volume='+0%', pitch='+0Hz', rate='+0%', speed=1.0, has_done=0, wait_sec=1.0, dub_nums=1, error='No audio file was generated during the listening test.', api_url='', is_cuda=False, local_dir='D:/AI_dub_pyvideotrans-v4.15/models/models--Qwen--Qwen3-TTS-12Hz-0.6B-CustomVoice', is_redubb=False, model_name='0.6B', ainame=None

=
system:Windows-10-10.0.26300-SP0
version:v4.15
frozen:True
language:en_US
root_dir:D:/AI_dub_pyvideotrans-v4.15

#1 DeepSeek reply at 2 hours ago

The Qwen3-TTS (Built-in) listening test failed to produce any audio. This almost always happens when the model files are incomplete/corrupted (the download from huggingface.co was interrupted), or when the reference audio used for voice cloning does not meet the required conditions.

Verify the model is complete. You are using the CustomVoice model. Check that BOTH of these folders exist and are fully populated (not partial downloads):
- D:/AI_dub_pyvideotrans-v4.15/models/models--Qwen--Qwen3-TTS-12Hz-0.6B-CustomVoice
- D:/AI_dub_pyvideotrans-v4.15/models/models--Qwen--Qwen3-TTS-12Hz-0.6B-Base
  If either looks incomplete, delete it and re-download it manually from:
  https://huggingface.co/Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice/tree/main
  https://huggingface.co/Qwen/Qwen3-TTS-12Hz-0.6B-Base/tree/main
Check the reference audio zh_male_bj.wav (used as the role). For Qwen3-TTS cloning it must be a single-channel WAV, 3–10 seconds long, clear single-speaker speech, with no long leading/trailing silence and no background noise. A bad reference clip is a very common cause of "no audio generated."
Confirm there is no invalid proxy configured (clear the proxy box in the Alignment row if you are not actually running a proxy), and that CUDA/Driver is compatible if is_cuda is enabled. If GPU errors occur, try disabling CUDA acceleration and rerun the listening test.
Make sure no other instance of sp.exe is running (they share the tmp folder and can conflict).

If the problem persists, please provide the full latest log from the logs/ folder (last ~30 lines) after retrying the listening test; the current trace only shows the symptom, not the underlying TTS sub-process error.

#7901 No audio file was generated during the listening test.self=[QwenttsLocal]: uuid=None, proxy_str=None, tts_type=1, len=1,

1 Replies

Please review the relevant documentation:

Post Your Reply

Related resource