onnx / onnx/models

转换v2模型后加载失败

Open
#685 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Jupyter Notebook
Stars
9.8k
Forks
1.6k
No merged PRs in 30d

Description

为什么将v2模型转换后，无法使用

Image Image 这是报错信息，这是因为转换有问题吗？ D:\Environment\python\python.exe F:\工程代码\TTS\tts.py Successfully generated temporary FP32 weights to improve inference speed. Error: Failed to load ONNX model 'F:\onxxx\t2s_encoder_fp32.onnx'. Details: [ONNXRuntimeError] : 1 : FAIL : Deserialize tensor encoder.ar_text_embedding.word_embeddings.weight failed.tensorprotoutils.cc:258 onnxruntime::utils::GetExternalDataInfo TensorProto: encoder.ar_text_embedding.word_embeddings.weight external data size mismatch. Computed size: 1499136, external_data.length: 1048576 Please call 'set_reference_audio' first to set the reference audio. 🎉 音频生成完毕!

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with tts.py and reproduce the failure while loading F:\onxxx\t2s_encoder_fp32.onnx. Inspect the generated ONNX external data for encoder.ar_text_embedding.word_embeddings.weight and compare its recorded length with the expected tensor size. Done means the converted v2 model loads without the external-data size mismatch.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.