run WuuShanghai and Cantonese error in KAN-TTS or ModelScope

[YangangCao]

python3.8 x86_64 Ubuntu server

run WuuShanghai in basemodel_16k directory in KAN-TTS: Traceback (most recent call last): File "kantts/bin/text_to_wav.py", line 218, in text_to_wav( File "kantts/bin/text_to_wav.py", line 157, in text_to_wav am_infer(symbols_file, am_ckpt, output_dir, se_file) File "/data2/caoyangang/open_project/KAN-TTS/kantts/bin/infer_sambert.py", line 217, in am_infer mel, mel_post, dur, f0, energy = am_synthesis( File "/data2/caoyangang/open_project/KAN-TTS/kantts/bin/infer_sambert.py", line 83, in am_synthesis inputs_ling = torch.stack( RuntimeError: stack expects each tensor to be equal size, but got [16] at entry 0 and [78] at entry 1

run WuuShanghai in voices directory in KAN-TTS get the same error as above

but run WuuShanghai in ModelScope is fine.

run Cantonese model in basemodel_16k directory in KAN-TTS: Traceback (most recent call last): File "kantts/bin/text_to_wav.py", line 218, in text_to_wav( File "kantts/bin/text_to_wav.py", line 157, in text_to_wav am_infer(symbols_file, am_ckpt, output_dir, se_file) File "/data2/caoyangang/open_project/KAN-TTS/kantts/bin/infer_sambert.py", line 217, in am_infer mel, mel_post, dur, f0, energy = am_synthesis( File "/data2/caoyangang/open_project/KAN-TTS/kantts/bin/infer_sambert.py", line 83, in am_synthesis inputs_ling = torch.stack( RuntimeError: stack expects each tensor to be equal size, but got [16] at entry 0 and [78] at entry 1

run Cantonese model in voices directory in KAN-TTS: Traceback (most recent call last): File "kantts/bin/text_to_wav.py", line 218, in text_to_wav( File "kantts/bin/text_to_wav.py", line 157, in text_to_wav am_infer(symbols_file, am_ckpt, output_dir, se_file) File "/data2/caoyangang/open_project/KAN-TTS/kantts/bin/infer_sambert.py", line 201, in am_infer fsnet.load_state_dict(state_dict["model"], strict=False) File "/home/eeodev/anaconda3/envs/kan-tts/lib/python3.8/site-packages/torch/nn/modules/module.py", line 2041, in load_state_dict raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format( RuntimeError: Error(s) in loading state_dict for KanTtsSAMBERT: size mismatch for text_encoder.sy_emb.weight: copying a param with shape torch.Size([107, 512]) from checkpoint, the shape in current model is torch.Size([147, 512]). size mismatch for text_encoder.tone_emb.weight: copying a param with shape torch.Size([14, 512]) from checkpoint, the shape in current model is torch.Size([10, 512]).

run Cantonese in ModelScope get the same error as above

[plutols]

@YangangCao Have you solved it?

[YangangCao]

don’t kantts for a long time, :)---- Replied Message @.>Date02/12/2025 11:18 @.> CcYangang @.>@.>SubjectRe: [modelscope/KAN-TTS] run WuuShanghai and Cantonese error in KAN-TTS or ModelScope (Issue #49) @YangangCao Have you solved it?

—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: @.> [ { @.": "http://schema.org", @.": "EmailMessage", "potentialAction": { @.": "ViewAction", "target": "https://github.com/modelscope/KAN-TTS/issues/49#issuecomment-2652564555", "url": "https://github.com/modelscope/KAN-TTS/issues/49#issuecomment-2652564555", "name": "View Issue" }, "description": "View this Issue on GitHub", "publisher": { @.***": "Organization", "name": "GitHub", "url": "https://github.com" } } ]

[YangangCao]

    don’t use kantts for a long time, :)---- Replied Message ***@***.***>Date02/12/2025 11:18 ***@***.***> CcYangang ***@***.***>***@***.***>SubjectRe: [modelscope/KAN-TTS] run WuuShanghai and Cantonese error in KAN-TTS or ModelScope (Issue #49)

@YangangCao Have you solved it?

—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: @.> [ { @.": "http://schema.org", @.": "EmailMessage", "potentialAction": { @.": "ViewAction", "target": "https://github.com/modelscope/KAN-TTS/issues/49#issuecomment-2652564555", "url": "https://github.com/modelscope/KAN-TTS/issues/49#issuecomment-2652564555", "name": "View Issue" }, "description": "View this Issue on GitHub", "publisher": { @.***": "Organization", "name": "GitHub", "url": "https://github.com" } } ]