Model Load Error

[Syobe]

.pthファイルを.onnxファイルにエクスポートした後、アップロードするとModel Load Errorが発生します Windows10 / VC Client v.1.5.2.8 / cpuでエクスポートしました AMD GPU/CPUを使用中です

[w-okada]

こちらの環境（Nvidia GPU+CPU）では再現しないですね・・。 エラーメッセージを貼りつけていただけますか。

[Syobe]

D:\VOICE\MMVCServerSIO>MMVCServerSIO.exe -p 18888 --https false --content_vec_500 checkpoint_best_legacy_500.pt --content_vec_500_onnx checkpoint_best_legacy_500.onnx --content_vec_500_onnx_on false --hubert_base hubert_base.pt --hubert_base_jp rinna_hubert_base_jp.pt --hubert_soft hubert-soft-0d54a1f4.pt --nsf_hifigan nsf_hifigan/model Booting PHASE :main Voice Changerを起動しています。 Internal_Port:18888 protocol: HTTP

---

ブラウザで次のURLを開いてください.
http://<IP>:<PORT>/
多くの場合は次のいずれかのURLにアクセスすると起動します。
http://localhost:18888/

Booting PHASE :__main__
Booting PHASE :MMVCServerSIO

VoiceChanger Initialized (GPU_NUM:0, mps_enabled:False) [2023-05-09 07:48:55] connet sid : _IJpjvJHcOu7q2M1AAAC [2023-05-09 07:48:55] connet sid : Ph5xciaJ0JiCCL3sAAAD 2023-05-09 07:48:56 | INFO | faiss.loader | Loading faiss with AVX2 support. 2023-05-09 07:48:56 | INFO | faiss.loader | Could not load library with AVX2 support due to: ModuleNotFoundError("No module named 'faiss.swigfaiss_avx2'") 2023-05-09 07:48:56 | INFO | faiss.loader | Loading faiss. 2023-05-09 07:48:56 | INFO | faiss.loader | Successfully loaded faiss. RVC initialization: VoiceChangerParams(content_vec_500='checkpoint_best_legacy_500.pt', content_vec_500_onnx='checkpoint_best_legacy_500.onnx', content_vec_500_onnx_on=0, hubert_base='hubert_base.pt', hubert_base_jp='rinna_hubert_base_jp.pt', hubert_soft='hubert-soft-0d54a1f4.pt', nsf_hifigan='nsf_hifigan/model') [Voice Changer] update configuration: dstId 1 [Voice Changer] update configuration: gpu -1 [Voice Changer] update configuration: crossFadeOffsetRate 0.1 CHAGE TO NEW PIPELINE False [Voice Changer] update configuration: crossFadeEndRate 1 [Voice Changer] update configuration: crossFadeOverlapSize 1024 [Voice Changer] update configuration: framework PyTorch [Voice Changer] update configuration: onnxExecutionProvider CPUExecutionProvider [Voice Changer] update configuration: f0Detector harvest [Voice Changer] update configuration: f0Factor 1 [Voice Changer] update configuration: silentThreshold 0.00001 f0Factor is not mutable variable or unknown variable! [Voice Changer] update configuration: extraConvertSize 32768 onnxExecutionProvider is not mutable variable or unknown variable! [Voice Changer] update configuration: clusterInferRatio 0.1 clusterInferRatio is not mutable variable or unknown variable! [Voice Changer] update configuration: indexRatio 1 [Voice Changer] update configuration: modelSamplingRate 48000 [Voice Changer] update configuration: inputSampleRate 48000 [Voice Changer] update configuration: silenceFront 1 [Voice Changer] update configuration: enableDirectML 1 [Voice Changer] RVC new model is uploaded,0 {'pyTorchModelFile': None, 'onnxModelFile': 'C:\Users\ADM\AppData\Local\Temp\tmplxr4qkoz\upload_dir\0\Sakura_voice_simple.onnx', 'featureFile': None, 'indexFile': None, 'defaultTrans': 0, 'isONNX': True, 'modelType': <EnumInferenceTypes.onnxRVCNono: 'onnxRVCNono'>, 'samplingRate': 48000, 'f0': False, 'embChannels': 256, 'deprecated': False, 'embedder': <EnumEmbedderTypes.hubert: 'hubert_base'>} [Voice Changer] Prepare Model of slot: 0 [Voice Changer] exception! loading inferencer 'OnnxRVCInferencerNono' object has no attribute 'onnx_model' Traceback (most recent call last): File "voice_changer\RVC\pipeline\PipelineGenerator.py", line 24, in createPipeline File "voice_changer\RVC\inferencer\InferencerManager.py", line 20, in getInferencer File "voice_changer\RVC\inferencer\InferencerManager.py", line 56, in loadInferencer File "voice_changer\RVC\inferencer\OnnxRVCInferencerNono.py", line 19, in loadModel AttributeError: 'OnnxRVCInferencerNono' object has no attribute 'onnx_model' [Voice Changer] generate new embedder. (no embedder) [Voice Changer] Loading index... Traceback (most recent call last): File "voice_changer\VoiceChanger.py", line 131, in loadModel File "voice_changer\RVC\RVC.py", line 86, in loadModel File "voice_changer\RVC\RVC.py", line 154, in prepareModel File "voice_changer\RVC\pipeline\PipelineGenerator.py", line 54, in createPipeline UnboundLocalError: local variable 'inferencer' referenced before assignment

[Voice Changer] Model Load Error! Check your model is valid. local variable 'inferencer' referenced before assignment

初期実行時にこのようなエラーも発生しますが、これが関連しているのでしょうか？

D:\VOICE\MMVCServerSIO>MMVCServerSIO.exe -p 18888 --https false --content_vec_500 checkpoint_best_legacy_500.pt --content_vec_500_onnx checkpoint_best_legacy_500.onnx --content_vec_500_onnx_on false --hubert_base hubert_base.pt --hubert_base_jp rinna_hubert_base_jp.pt --hubert_soft hubert-soft-0d54a1f4.pt --nsf_hifigan nsf_hifigan/model Booting PHASE :main Voice Changerを起動しています。 Internal_Port:18888 protocol: HTTP

---

ブラウザで次のURLを開いてください.
http://<IP>:<PORT>/
多くの場合は次のいずれかのURLにアクセスすると起動します。
http://localhost:18888/

Booting PHASE :__main__
Booting PHASE :MMVCServerSIO

55%|██████████████████████████████████████████▌ | 98.6M/181M [00:01<00:01, 67.6MB/s]did-fail-load█▍ | 39.2M/361M [00:01<00:07, 43.9MB/s] (node:12524) electron: Failed to load URL: http://localhost:18888/ with error: ERR_CONNECTION_REFUSED (Use voice-changer-native-client --trace-warnings ... to show where the warning was created) did-fail-load███████████████████████████████████████████████████████▎ | 281M/361M [00:04<00:00, 111MB/s] (node:12524) electron: Failed to load URL: http://localhost:18888/ with error: ERR_CONNECTION_REFUSED VoiceChanger Initialized (GPU_NUM:0, mps_enabled:False)

[w-okada]

ありがとうございます。f0のモデルですね。再現しました！

[w-okada]

v.1.5.2.9で修正が入りました。ご確認ください。

[Syobe]

確認してみたら正常に作動します！ ありがとうございました！ DirectML対応も期待します :D

[chuan6v]

ブラウザで次のURLを開いてください. https://:/ 多くの場合は次のいずれかのURLにアクセスすると起動します。 https://localhost:18888/ https://192.168.2.19:18888/ Booting PHASE :MMVCServerSIO VoiceChanger Initialized (GPU_NUM:1, mps_enabled:False) 这个是什么意思