ErrLookup › microsoft/VibeVoice
microsoft/VibeVoice
Open-Source Frontier Voice AI · Python · 35 source files
Analyzed at 94da20d98b on 2026-08-15. 61 documented errors.
| Code / Message | Type | Severity | Tags |
|---|---|---|---|
| Missing acoustic/semantic tokenizer config in model config | exception | critical | model-config, checkpoint, huggingface, initialization, vllm |
| Audio at index is too short to be represented | exception | error | audio, batching, prompt-construction, data-quality, vllm |
| Voice preset not found | exception | error | web-demo, voice-preset, invalid-key, runtimeerror |
| missing `sample` as a required keyword argument | exception | error | python, diffusion, scheduler, internal-api, api-misuse |
| Multiple voice presets match the speaker name | validation | error | demo, voice-preset, ambiguous-match, valueerror |
| is not implemented for | exception | error | python, diffusion, scheduler, config, not-implemented |
| load_audio_bytes_use_ffmpeg requires resample=True | validation | error | audio, ffmpeg, resampling, io |
| prediction_type given as | exception | error | python, diffusion, scheduler, config, prediction-type |
| Voices directory not found | exception | critical | web-demo, startup, missing-directory, voice-preset, environment |
| No valid entries found in JSON file | validation | error | json, script, validation, data-quality |
| `final_sigmas_type` must be one of 'zero', or 'sigma_min'… | exception | error | python, diffusion, scheduler, config, sigma |
| acoustic_tokenizer_config has unexpected type | exception | critical | model-config, type-error, serialization, omegaconf, vllm |
| Unsupported alpha_transform_type | exception | error | python, diffusion, scheduler, config, noise-schedule |
| Unsupported audio type | exception | error | python, audio, type-validation, input-contract |
| ffmpeg returned empty audio data | exception | error | ffmpeg, subprocess, audio-decode, demo-script, media |
| soundfile is required to save audio files. Install it with… | exception | error | python, audio, dependency, optional-import, environment |
| Failed to load audio | exception | error | ffmpeg, error-wrapping, audio-decode, demo-script |
| is not supported. Please make sure to choose one of… | exception | error | python, diffusion, scheduler, config, timesteps |
| Unsupported audio data type | exception | error | audio, input-validation, multimodal, vllm, type-error |
| Unexpected audio shape | exception | error | audio, shape, mono, input-validation |
| Unsupported tokenizer type for | validation | error | tokenizer, config, from-pretrained, misleading-error |
| Sample index exceeds batch size | validation | error | streaming, indexing, batch, audio |
| Please install huggingface_hub: pip install huggingface_hub | exception | warning | dependencies, import-error, huggingface, tokenizer, tooling |
| missing`sample` as a required keyword argument | exception | error | python, diffusion, scheduler, internal-api, api-misuse |
| Cannot find embed_tokens layer | exception | critical | vllm, version-compatibility, embedding, model-wrappers, attribute-error |
| `final_sigmas_type` is not supported for `algorithm_type` … | exception | error | python, diffusion, scheduler, config, sigma |
| Unsupported dist_type | validation | error | sampling, config, tokenizer, vae |
| Unsupported tokenizer type for | validation | error | tokenizer, config, asr, from-pretrained |
| VibeVoiceStreamingProcessor.__call__ is not implemented… | exception | error | streaming, api-misuse, not-implemented, processor |
| No valid speaker lines found in script | validation | error | script, parsing, input-format, tts |
| Cannot use `timesteps` with `config.use_karras_sigmas =… | exception | error | python, diffusion, scheduler, config, karras, timesteps |
| Number of inference steps is 'None', you need to run… | exception | error | python, diffusion, scheduler, api-misuse, lifecycle |
| Unsupported decoder model type | validation | error | configuration, decoder, qwen2, valueerror |
| Audio duration ( s) exceeds the configured limit ( s). Set… | exception | error | audio, duration-limit, oom-prevention, configuration, vllm |
| Audio should be 1D or 2D, got shape | exception | error | audio, shape, ndim, input-validation |
| Unsupported file format | exception | error | audio, file-format, ffmpeg, validation |
| Could not process input text | validation | error | input-validation, script, text-processing, tts |
| segment_length must be positive | validation | error | asr, streaming, segment-duration, valueerror, parameter-validation |
| VibeVoiceStreamingModel.forward is intentionally disabled… | exception | error | api-contract, streaming, forward-disabled, runtimeerror |
| Unified forward is disabled. Use `forward_lm`… | exception | error | api-contract, streaming, inference, forward-disabled, runtimeerror |
| missing `sample` as a required keyword argument | exception | error | python, diffusion, scheduler, internal-api, api-misuse |
| Must pass exactly one of `num_inference_steps` or… | exception | error | python, diffusion, scheduler, api-misuse, timesteps |
| No voice preset (.pt) files found in | exception | critical | web-demo, startup, voice-preset, empty-directory, download |
| Output embeddings (lm_head) are not defined for this model… | exception | error | api-contract, streaming, lm-head, embeddings, runtimeerror |
| Can only pass one of `num_inference_steps` or… | exception | error | python, diffusion, scheduler, api-misuse, timesteps |
| Cannot use `timesteps` with `config.use_lu_lambdas = True` | exception | error | python, diffusion, scheduler, config, lu-lambdas, timesteps |
| Unsupported mixer layer | validation | error | tokenizer, mixer-layer, configuration, valueerror |
| Unsupported modality | exception | error | modality, multimodal, prompt-template, vllm, routing |
| Unsupported norm type | validation | error | tokenizer, layernorm, configuration, valueerror |
| Prediction type not implemented | exception | error | training, diffusion, prediction-type, config, notimplementederror |
| Speech type not implemented | exception | error | training, collator, speech-type, notimplementederror |
| No valid content found in text file | validation | error | text-file, script, validation, data-quality |
| GroupNorm doesn't support causal evaluation. | validation | error | tokenizer, normalization, causal, groupnorm, valueerror |
| JSON file must contain a list of speaker entries | validation | error | json, script, file-format, validation |
| Audio input is required | exception | error | audio, validation, input-validation, tokenizer |
| Empty audio list provided | exception | error | audio, batch, validation, edge-case |
| MODEL_PATH not set in environment | exception | critical | web-demo, environment, configuration, startup |
| semantic_tokenizer_config has unexpected type | exception | critical | model-config, type-error, serialization, vllm |
| Loss computation is not implemented in this version. | exception | error | streaming, inference, labels, notimplementederror, training-unsupported |
| StreamingTTSService not initialized | exception | error | web-demo, lifecycle, not-initialized, race-condition |
| Audio input is required for ASR processing | validation | error | asr, validation, audio, input-validation |