ErrLookup › RVC-Boss/GPT-SoVITS

RVC-Boss/GPT-SoVITS

1 min voice data can also be used to train a good TTS model! (few shot voice cloning) · Python · 153 source files

Analyzed at d523079fc0 on 2026-08-15. 21 documented errors.

Code / MessageTypeSeverityTags
SoVITS %s 底模缺失，无法加载相应 LoRA 权重
exception error model-weights, lora, sovits, missing-file, initialization
参考音频在3~10秒范围外，请更换！
validation error reference-audio, audio-duration, validation, tts
prompt_text cannot be empty when using SoVITS_V3
validation error prompt-text, v2pro, vocoder, tts, validation
ref_audio_path cannot be empty, when the reference audio is
validation error reference-audio, validation, tts, api-misuse
{ref_audio_path} not exists
validation error file-path, reference-audio, validation, tts
SoVits V3/4模型不支持流式推理模式
exception error streaming, sovits, v3, v4, model-capability
请输入有效文本
validation error text-validation, empty-input, tts, preprocessing
{path_sovits}SoVITS {model_version}底模缺失，无法加载相应 LoRA 权重
exception error model-weights, lora, sovits, webui, missing-file
参考音频在3~10秒范围外，请更换！
validation error reference-audio, audio-duration, webui, validation
CUDA Graph T2S inference failed
exception error cuda, cuda-graph, t2s, gpu, inference
请输入有效文本
validation error text-validation, empty-input, webui, preprocessing
{path_sovits}SoVITS {model_version}底模缺失，无法加载相应 LoRA 权重
exception error model-weights, lora, sovits, webui-fast, missing-file
搜索音频长度 ({search_len}) 必须大于等于参考音频长度 ({ref_len})
validation error streaming, audio-alignment, shape-validation, v2pro
Cannot locate {filename} in candidate dirs: {candidate_dirs}
exception error g2pw, model-weights, missing-file, chinese-frontend, onnx
Files not found: {paths}
exception error g2pw, model-weights, missing-file, onnx, chinese-frontend
{language} is not supported. Supported: zh, yue, ja, en, ko,
validation error asr, funasr, language-code, validation
You input a wrong audio path that does not exists, please fi
validation error audio-io, file-path, training-data, validation
音频加载失败
exception error audio-io, ffmpeg, environment, training-data
The following condition must be satisfied: min_length >= min
validation error audio-slicing, parameter-validation, config, preprocessing
The following condition must be satisfied: max_sil_kept >= h
validation error audio-slicing, parameter-validation, config, preprocessing
Error: Unknown model type. If you are using a model without
validation error uvr5, audio-separation, model-loading, configuration, filename-convention