#7581 The voice actor has been cloned, but there is no corresponding reference audio. Please choose a different voice actor.Tr

2402**9475 Posted at: 1 hour ago

The voice actor has been cloned, but there is no corresponding reference audio. Please choose a different voice actor.
Traceback (most recent call last):
File "videotrans\task\only_one.py", line 107, in run
File "videotrans\task\_stage_dubbing.py", line 24, in dubbing
File "videotrans\task\_stage_dubbing.py", line 101, in _tts
File "videotrans\tts\__init__.py", line 118, in run
File "videotrans\tts\_base.py", line 135, in run
File "videotrans\tts\_base.py", line 213, in _item_task
File "videotrans\tts\_fishtts.py", line 24, in _run
File "videotrans\tts\_base.py", line 291, in get_ref_wav
RuntimeError: The voice actor has been cloned, but there is no corresponding reference audio. Please choose a different voice actor.

cfg=[TaskCfgVTT]当前工作模式: 翻译视频单视频模式
原始输入文件名: D:/YOUTUBE/ARROW/TACH AM/1 VOICE SACH.mp4,
输出结果保存到文件夹: D:/YOUTUBE/ARROW/TACH AM/_video_out/1 VOICE SACH-mp4,
临时文件夹: D://tmp/7132/64de268f27
已选清理已存在
已启用CUDA加速
未选降噪
已选识别说话人，最大说话人数量不限制
STT Service:faster-whisper(Built-in), large-v3-turbo, 发音语言: English, 断句方式:默认断句
发音语言和目标语言相同，不翻译字幕
TTS Service:Fish TTS(LocalAPI), 角色:clone, 配音语言:English, 未选二次语音识别
音量:+5%, 语速:-15%, 已选音频加速, 已选视频慢速
字幕: Hard Subtitles
已选分离人声与背景声, 已选重新嵌入背景声, 背景音量0.8, 背景声音时长短于视频时长时: 拉长(降速播放),存在分离后的纯净人声文件,存在分离后的背景声音文件
video_duration_ms=2007400

=
system:Windows-10-10.0.26200-SP0
version:v4.13
frozen:True
language:en_US
root_dir:D:/

#1 DeepSeek reply at 1 hour ago

The clone voice role requires the Fish TTS (LocalAPI) channel to support voice cloning by receiving a reference audio file. Fish TTS as configured does not accept a clone role — it has no matching reference-audio entry, so get_ref_wav() raised this error.

Solutions (choose one):

Switch to a TTS channel that supports clone (recommended if you need voice cloning):
Channels supporting clone: Qwen3-TTS, F5-TTS, OmniVoice-TTS, Confucius-TTS, MOSS-TTS-Nano, ZipVoice, ChatterBox, Higgs-audio-v3, GPT-SoVITS, Index-TTS, VoxCPM-TTS, CosyVoice, Spark-TTS.
Or, if you want to keep Fish TTS: change the 配音角色 (Voice Role) from clone to a specific built-in voice listed for Fish TTS.
Ensure clone prerequisites are met (required for good clone results regardless of channel):
- Menu -> Tools -> Advanced Options -> ASR Settings: set 最短语音持续(毫秒) to 3000–4000 (must be ≥3000 for clone), and 最长语音持续(秒) to 6–10.
- Reference audio must be 3–10s; fragments cut from the original video shorter than 3s cannot be used as reference audio and trigger this kind of error.

Note: Fish-TTS is a LocalAPI channel and does not use clone in the same way as built-in clone channels — in the current channel list it is not marked as supporting the clone role. For cloning, use one of the channels in item 1.

#7581 The voice actor has been cloned, but there is no corresponding reference audio. Please choose a different voice actor.Tr

1 Replies

Please check the related docs:

Post Your Reply

Related resource