search

shaojinding / Adversarial-Many-to-Many-VC

[InterSpeech 2020] "Improving the Speaker Identity of Non-Parallel Many-to-Many VoiceConversion with Adversarial Speaker Recognition" by Shaojin Ding, Guanlong Zhao, Ricardo Gutierrez-Osuna
Other
39 stars 12 forks source link

Error while running synthesis_ppg_script.py #6

Open narendranp opened 3 years ago

narendranp commented 3 years ago

Hi,

I am getting following the error while trying to run synthesis_ppg_script.py. I am using the following command: python synthesis_ppg_script.py /home/ubuntu/narendra/VC_dataset/SV2TTS/synthesizer_test/ /home/ubuntu/narendra/Adversarial-Many-to-Many-VC-master/ --num_speakers 2 --num_utterances 2

Also, can please let me know what are the changes to be done in hparams.py or any other file for running synthesis_ppg_script.py.

Traceback (most recent call last): File "synthesis_ppg_script.py", line 138, in main() File "synthesis_ppg_script.py", line 132, in main generated_wavs_avg = synthesize_ppg_batch_avg_embed(ppg_paths, embed_paths) File "synthesis_ppg_script.py", line 49, in synthesize_ppg_batch_avg_embed generated_wavs.append(synthesize_ppg(ppg, avg_embed)) File "synthesis_ppg_script.py", line 22, in synthesize_ppg specs = synthesizer.synthesize_spectrograms([ppg], [embed]) File "/home/ubuntu/narendra/Adversarial-Many-to-Many-VC-master/synthesizer/inference.py", line 80, in synthesize_spectrograms self.load() File "/home/ubuntu/narendra/Adversarial-Many-to-Many-VC-master/synthesizer/inference.py", line 61, in load self._model = Tacotron2(self.checkpoint_fpath, hparams) File "/home/ubuntu/narendra/Adversarial-Many-to-Many-VC-master/synthesizer/tacotron2.py", line 28, in init split_infos=split_infos) File "/home/ubuntu/narendra/Adversarial-Many-to-Many-VC-master/synthesizer/models/tacotron.py", line 178, in initialize p = tf.cast(global_step, tf.float32) / tf.cast(tacotron_train_steps, tf.float32) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/util/dispatch.py", line 180, in wrapper return target(*args, **kwargs) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/ops/math_ops.py", line 614, in cast x = ops.convert_to_tensor(x, name="x") File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 1039, in convert_to_tensor return convert_to_tensor_v2(value, dtype, preferred_dtype, name) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 1097, in convert_to_tensor_v2 as_ref=False) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 1175, in internal_convert_to_tensor ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/constant_op.py", line 304, in _constant_tensor_conversion_function return constant(v, dtype=dtype, name=name) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/constant_op.py", line 245, in constant allow_broadcast=True) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/constant_op.py", line 283, in _constant_impl allow_broadcast=allow_broadcast)) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/tensor_util.py", line 454, in make_tensor_proto raise ValueError("None values not supported.") ValueError: None values not supported.

shaojinding commented 3 years ago

Hi,

I think it should be a TF version issue. All my code was tested with TF 13.1. Which TF version are you using? Could you please try 13.1?

On Fri, Jul 16, 2021 at 00:12 narendranp @.***> wrote:

Hi,

I am getting following the error while trying to run synthesis_ppg_script.py. I am using the following command: python synthesis_ppg_script.py /home/ubuntu/narendra/VC_dataset/SV2TTS/synthesizer_test/ /home/ubuntu/narendra/Adversarial-Many-to-Many-VC-master/ --num_speakers 2 --num_utterances 2

Also, can please let me know what are the changes to be done in hparams.py or any other file for running synthesis_ppg_script.py.

Traceback (most recent call last): File "synthesis_ppg_script.py", line 138, in main() File "synthesis_ppg_script.py", line 132, in main generated_wavs_avg = synthesize_ppg_batch_avg_embed(ppg_paths, embed_paths) File "synthesis_ppg_script.py", line 49, in synthesize_ppg_batch_avg_embed generated_wavs.append(synthesize_ppg(ppg, avg_embed)) File "synthesis_ppg_script.py", line 22, in synthesize_ppg specs = synthesizer.synthesize_spectrograms([ppg], [embed]) File "/home/ubuntu/narendra/Adversarial-Many-to-Many-VC-master/synthesizer/inference.py", line 80, in synthesize_spectrograms self.load() File "/home/ubuntu/narendra/Adversarial-Many-to-Many-VC-master/synthesizer/inference.py", line 61, in load self._model = Tacotron2(self.checkpoint_fpath, hparams) File "/home/ubuntu/narendra/Adversarial-Many-to-Many-VC-master/synthesizer/tacotron2.py", line 28, in init split_infos=split_infos) File "/home/ubuntu/narendra/Adversarial-Many-to-Many-VC-master/synthesizer/models/tacotron.py", line 178, in initialize p = tf.cast(global_step, tf.float32) / tf.cast(tacotron_train_steps, tf.float32) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/util/dispatch.py", line 180, in wrapper return target(*args, **kwargs) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/ops/math_ops.py", line 614, in cast x = ops.convert_to_tensor(x, name="x") File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 1039, in convert_to_tensor return convert_to_tensor_v2(value, dtype, preferred_dtype, name) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 1097, in convert_to_tensor_v2 as_ref=False) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/ops.py", line 1175, in internal_convert_to_tensor ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/constant_op.py", line 304, in _constant_tensor_conversion_function return constant(v, dtype=dtype, name=name) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/constant_op.py", line 245, in constant allow_broadcast=True) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/constant_op.py", line 283, in _constant_impl allow_broadcast=allow_broadcast)) File "/home/ubuntu/anaconda3/envs/vc-speech/lib/python3.7/site-packages/tensorflow/python/framework/tensor_util.py", line 454, in make_tensor_proto raise ValueError("None values not supported.") ValueError: None values not supported.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://urldefense.com/v3/__https://github.com/shaojinding/Adversarial-Many-to-Many-VC/issues/6__;!!KwNVnqRv!SrOnCHZYM0n9-PJfXiWWa4rztbA3QEBNTP8yZnzYH6ykRx8TqRuZmWI9zyrVJw$, or unsubscribe https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/AFC3LL7ZE5OHJRMQVWYMFZTTX65UJANCNFSM5AO3MV5Q__;!!KwNVnqRv!SrOnCHZYM0n9-PJfXiWWa4rztbA3QEBNTP8yZnzYH6ykRx8TqRuZmWIwKNeghA$ .

-- Shaojin Ding, Ph.D. in Computer Science, Texas A&M University

narendranp commented 3 years ago

Hi, Thanks for the reply. I am using tensorflow-gpu=1.13.1. I was getting similar kind of error while training. During that time in hparams.py, I changed "if_use_speaker_classifier=True", "n_speakers=4(no. spkrs used)", then error had gone. So, I thought there is something I am missing (e.g. changing some parameters in hparams.py or other files). Is the setup used for training and testing is same or there any changes.