ErrLookup › OpenBMB/VoxCPM
OpenBMB/VoxCPM
VoxCPM2: Tokenizer-Free TTS for Multilingual Speech Generation, Creative Voice Design, and True-to-Life Cloning · Python · 45 source files
Analyzed at f5a1c6a6b9 on 2026-08-27. 20 documented errors.
| Code / Message | Type | Severity | Tags |
|---|---|---|---|
| Retry on bad cases is not supported in streaming mode… | console | warning | voxcpm, voxcpm2, tts, streaming, retry, configuration, user-warning |
| ' ' does not exist | validation | error | cli, file-not-found, path-validation |
| Timestamp alignment failed | exception | error | cli, timestamp-alignment, strict-mode |
| prompt_wav_path and prompt_text must both be provided or… | validation | error | voice-cloning, parameter-validation, paired-arguments |
| Unsupported architecture | validation | error | model-config, architecture, value-error |
| target text must be a non-empty string | validation | error | input-validation, tts, empty-input |
| Requested device ' ', but CUDA is not available. Use… | validation | error | cuda, device-selection, environment |
| Expected string input, got | validation | error | tokenizer, type-error, input-validation |
| VoxCPMModel can only be optimized on CUDA device | validation | error | optimization, torch-compile, cuda-only |
| VOXCPM_MPS_DTYPE=' ' is not one of | validation | error | mps, dtype, environment-variable, macos |
| Cannot load LoRA weights: model was not initialized with… | exception | error | lora, fine-tuning, model-state |
| Tokenization failed | exception | error | tokenizer, error-chaining, value-error |
| Unsupported device ' '. Supported values are 'auto', 'cpu'… | validation | error | device-selection, value-error, configuration |
| triton is not installed | validation | error | triton, missing-dependency, optimization |
| Requested device 'mps', but MPS is not available. Use… | validation | error | mps, device-selection, macos |
| Unsupported dtype | validation | error | dtype, configuration, value-error |
| You must provide hf_model_id | validation | error | model-loading, missing-argument, huggingface |
| reference_wav_path is only supported with VoxCPM2 models | validation | error | version-mismatch, voice-cloning, feature-support |
| prompt_wav_path does not exist | validation | error | voice-cloning, file-not-found, prompt-audio |
| reference_wav_path does not exist | validation | error | voice-cloning, file-not-found, reference-audio |