2025-03-11 22:26:11,569 - modelscope - INFO - PyTorch version 2.4.1+cu118 Found.
14
2025-03-11 22:26:11,570 - modelscope - INFO - Loading ast index from E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_cache\modelscope\ast_indexer
15
2025-03-11 22:26:11,992 - modelscope - INFO - Loading done! Current index file version is 1.15.0, with md5 f5e62fd2891b889c4412a46d09dad12e and a total number of 980 components indexed
16
transformer is not installed, please install it if you want to use related modules
17
failed to import ttsfrd, use WeTextProcessing instead
18
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\diffusers\models\lora.py:393: FutureWarning: `LoRACompatibleLinear` is deprecated and will be removed in version 1.0.0. Use of `LoRACompatibleLinear` is deprecated. Please switch to PEFT backend by installing PEFT: `pip install peft`.
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\torch\nn\utils\weight_norm.py:134: FutureWarning: `torch.nn.utils.weight_norm` is deprecated in favor of `torch.nn.utils.parametrizations.weight_norm`.
22
WeightNorm.apply(module, name, dim)
23
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\cosyvoice\dataset\processor.py:24: UserWarning: torchaudio._backend.set_audio_backend has been deprecated. With dispatcher enabled, this function is no-op. You can remove the function call.
24
torchaudio.set_audio_backend('soundfile')
25
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py:69: UserWarning: Specified provider 'CUDAExecutionProvider' is not in available provider names.Available providers: 'AzureExecutionProvider, CPUExecutionProvider'
26
warnings.warn(
27
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\cosyvoice\cli\frontend.py:57: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
2025-03-11 22:26:23,603 WETEXT INFO found existing fst: E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\zh_tn_tagger.fst
30
2025-03-11 22:26:23,603 INFO found existing fst: E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\zh_tn_tagger.fst
31
2025-03-11 22:26:23,603 WETEXT INFO E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\zh_tn_verbalizer.fst
32
2025-03-11 22:26:23,603 INFO E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\zh_tn_verbalizer.fst
33
2025-03-11 22:26:23,603 WETEXT INFO skip building fst for zh_normalizer ...
34
2025-03-11 22:26:23,603 INFO skip building fst for zh_normalizer ...
35
2025-03-11 22:26:23,964 WETEXT INFO found existing fst: E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\en_tn_tagger.fst
36
2025-03-11 22:26:23,964 INFO found existing fst: E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\en_tn_tagger.fst
37
2025-03-11 22:26:23,964 WETEXT INFO E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\en_tn_verbalizer.fst
38
2025-03-11 22:26:23,964 INFO E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\en_tn_verbalizer.fst
39
2025-03-11 22:26:23,964 WETEXT INFO skip building fst for en_normalizer ...
40
2025-03-11 22:26:23,964 INFO skip building fst for en_normalizer ...
41
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\cosyvoice\cli\model.py:57: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\cosyvoice\cli\model.py:60: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\cosyvoice\cli\model.py:62: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
模型日志看一下，在提交工单的时候可以自动收集模型日志
@aigcpanel:
请问是这个么？
软件和模型都升级到最新的版本
time="2025-03-11T22:25:53+08:00" level=info msg="Failed to kill existing process:failed to find process: OpenProcess: The parameter is incorrect."
1
time="2025-03-11T22:25:53+08:00" level=info msg="server starting" buildTime="2025-02-24 12:20:07" debug=false version=0.0.1
2
time="2025-03-11T22:25:54+08:00" level=info msg="start api mode" port=50618
3
################## AigcPanelServer ###################
4
root: E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0
5
platformName: win
6
platformArch: x86
7
name: cosyvoice
8
######################################################
9
2025-03-11 22:26:08 - 开始运行
10
2025-03-11 22:26:08 - config {'id': 'SoundClone_1', 'mode': 'local', 'modelConfig': {'type': 'soundClone', 'seed': 0, 'speed': 1, 'text': '春天在哪里', 'promptAudio': 'C:\\Users\\shenx\\AppData\\Roaming\\aigcpanel\\data\\soundClonePrompt\\aj5nJ8PpTGEDZigj.wav', 'promptText': '秋天到了，中医怎么养生，记住这几点，这样冬天养生简单又实用，身体要养的好，心情才舒畅，第一适当温补冬季人体阳气内藏', 'crossLingual': False}, 'setting': {}} ['E:\\aigcpanel-server-cosyvoice-win-x86-v0.3.0\\aigcpanelrun.py', 'C:\\Users\\shenx\\AppData\\Roaming\\aigcpanel\\temp\\file_1741703167061_OOiLkKsYx9jdiDFVIOJwefmz3WB7X0J3.json']
11
Result[SoundClone_1][{"UseCuda": true}]
12
AigcPanelRunResult[SoundClone_1][eyJVc2VDdWRhIjogdHJ1ZX0=]
13
2025-03-11 22:26:11,569 - modelscope - INFO - PyTorch version 2.4.1+cu118 Found.
14
2025-03-11 22:26:11,570 - modelscope - INFO - Loading ast index from E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_cache\modelscope\ast_indexer
15
2025-03-11 22:26:11,992 - modelscope - INFO - Loading done! Current index file version is 1.15.0, with md5 f5e62fd2891b889c4412a46d09dad12e and a total number of 980 components indexed
16
transformer is not installed, please install it if you want to use related modules
17
failed to import ttsfrd, use WeTextProcessing instead
18
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\diffusers\models\lora.py:393: FutureWarning: `LoRACompatibleLinear` is deprecated and will be removed in version 1.0.0. Use of `LoRACompatibleLinear` is deprecated. Please switch to PEFT backend by installing PEFT: `pip install peft`.
19
deprecate("LoRACompatibleLinear", "1.0.0", deprecation_message)
20
2025-03-11 22:26:20,729 INFO input frame rate=50
21
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\torch\nn\utils\weight_norm.py:134: FutureWarning: `torch.nn.utils.weight_norm` is deprecated in favor of `torch.nn.utils.parametrizations.weight_norm`.
22
WeightNorm.apply(module, name, dim)
23
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\cosyvoice\dataset\processor.py:24: UserWarning: torchaudio._backend.set_audio_backend has been deprecated. With dispatcher enabled, this function is no-op. You can remove the function call.
24
torchaudio.set_audio_backend('soundfile')
25
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py:69: UserWarning: Specified provider 'CUDAExecutionProvider' is not in available provider names.Available providers: 'AzureExecutionProvider, CPUExecutionProvider'
26
warnings.warn(
27
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\cosyvoice\cli\frontend.py:57: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
28
self.spk2info = torch.load(spk2info, map_location=self.device)
29
2025-03-11 22:26:23,603 WETEXT INFO found existing fst: E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\zh_tn_tagger.fst
30
2025-03-11 22:26:23,603 INFO found existing fst: E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\zh_tn_tagger.fst
31
2025-03-11 22:26:23,603 WETEXT INFO E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\zh_tn_verbalizer.fst
32
2025-03-11 22:26:23,603 INFO E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\zh_tn_verbalizer.fst
33
2025-03-11 22:26:23,603 WETEXT INFO skip building fst for zh_normalizer ...
34
2025-03-11 22:26:23,603 INFO skip building fst for zh_normalizer ...
35
2025-03-11 22:26:23,964 WETEXT INFO found existing fst: E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\en_tn_tagger.fst
36
2025-03-11 22:26:23,964 INFO found existing fst: E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\en_tn_tagger.fst
37
2025-03-11 22:26:23,964 WETEXT INFO E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\en_tn_verbalizer.fst
38
2025-03-11 22:26:23,964 INFO E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\_aienv\lib\site-packages\tn\en_tn_verbalizer.fst
39
2025-03-11 22:26:23,964 WETEXT INFO skip building fst for en_normalizer ...
40
2025-03-11 22:26:23,964 INFO skip building fst for en_normalizer ...
41
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\cosyvoice\cli\model.py:57: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
42
self.llm.load_state_dict(torch.load(llm_model, map_location=self.device))
43
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\cosyvoice\cli\model.py:60: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
44
self.flow.load_state_dict(torch.load(flow_model, map_location=self.device))
45
E:\aigcpanel-server-cosyvoice-win-x86-v0.3.0\cosyvoice\cli\model.py:62: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
46
self.hift.load_state_dict(torch.load(hift_model, map_location=self.device))
47
2025-03-11 22:26:34.062869 - soundClone {'type': 'soundClone', 'seed': 0, 'speed': 1, 'text': '春天在哪里', 'promptAudio': 'C:\\Users\\shenx\\AppData\\Roaming\\aigcpanel\\data\\soundClonePrompt\\aj5nJ8PpTGEDZigj.wav', 'promptText': '秋天到了，中医怎么养生，记住这几点，这样冬天养生简单又实用，身体要养的好，心情才舒畅，第一适当温补冬季人体阳气内藏', 'crossLingual': False}
48
49
0%| | 0/1 [00:00<?, ?it/s]2025-03-11 22:26:45,977 INFO synthesis text 春天在哪里。
我看了一下 软件和模型都是最新的
你提交一个工单，收集一下完整日志我们看一下