I can't run it in window 10, could someone help me ?

[flyuuo9]

My env is win10 + anaconda2 + python3.5. It's my first time to use tensorflow. The log below looks like something went wrong when parse hparams/default.yaml. I even have tried changed default.yaml the CF to window's CRLF. Cound someone help me ?

(python35) λ pip show pyyaml
Name: PyYAML
Version: 3.13
Summary: YAML parser and emitter for Python
Home-page: http://pyyaml.org/wiki/PyYAML
Author: Kirill Simonov
Author-email: xi@resolvent.net
License: MIT

(python35) λ pip show tensorflow
Name: tensorflow
Version: 1.9.0
Summary: TensorFlow is an open source machine learning framework for everyone.
Home-page: https://www.tensorflow.org/
Author: Google Inc.
Author-email: opensource@google.com
License: Apache 2.0

D:\proj_github\deep-voice-conversion (master -> origin)
(python35) λ python train1.py case
case: case, logdir: /data/private/vc/logdir/case/train1
[0725 16:52:49 @logger.py:109] WRN Log directory /data/private/vc/logdir/case/train1 exists! Use 'd' to delete it.
[0725 16:52:49 @logger.py:112] WRN If you're resuming from a previous run, you can choose to keep it.
Press any other key to exit.
Select Action: k (keep) / d (delete) / q (quit):d
[0725 16:52:52 @logger.py:74] Argv: train1.py case
[0725 16:52:52 @parallel.py:175] WRN MultiProcessPrefetchData does support windows. However, windows requires more strict picklability on processes, which may lead of failure on some of the code.
[0725 16:52:52 @parallel.py:185] [MultiProcessPrefetchData] Will fork a dataflow more than one times. This assumes the datapoints are i.i.d.
Process _Worker-1:
Traceback (most recent call last):
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\multiprocessing\process.py", line 252, in _bootstrap
    self.run()
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorpack\dataflow\parallel.py", line 162, in run
    for dp in self.ds.get_data():
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorpack\dataflow\common.py", line 116, in get_data
    for data in self.ds.get_data():
  File "D:\proj_github\deep-voice-conversion\data_load.py", line 35, in get_data
    yield get_mfccs_and_phones(wav_file=wav_file)
  File "D:\proj_github\deep-voice-conversion\data_load.py", line 72, in get_mfccs_and_phones
    wav = read_wav(wav_file, sr=hp.default.sr)
KeyError: 'default'
Process _Worker-2:
Traceback (most recent call last):
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\multiprocessing\process.py", line 252, in _bootstrap
    self.run()
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorpack\dataflow\parallel.py", line 162, in run
    for dp in self.ds.get_data():
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorpack\dataflow\common.py", line 116, in get_data
    for data in self.ds.get_data():
  File "D:\proj_github\deep-voice-conversion\data_load.py", line 35, in get_data
    yield get_mfccs_and_phones(wav_file=wav_file)
  File "D:\proj_github\deep-voice-conversion\data_load.py", line 72, in get_mfccs_and_phones
    wav = read_wav(wav_file, sr=hp.default.sr)
KeyError: 'default'

[0725 16:52:31 @training.py:101] Building graph for training tower 1 on device /gpu:1 ...
[0725 16:52:34 @collection.py:164] These collections were modified but restored in tower1: (tf.GraphKeys.SUMMARIES: 3->5)
Traceback (most recent call last):
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorflow\python\framework\ops.py", line 1589, in _create_c_op
    c_op = c_api.TF_FinishOperation(op_desc)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Op type not registered 'NcclAllReduce' in binary running on mywind-PC. Make sure the Op and Kernel are registered in the binary running in this process. Note that if you are loading a saved graph which used ops from tf.contrib, accessing (e.g.) `tf.contrib.resampler` should be done before importing the graph, as contrib ops are lazily registered when the module is first accessed. while building NodeDef 'AllReduceGrads/NcclAllReduce'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "D:/proj_github/deep-voice-conversion/train1.py", line 78, in <module>
    train(args, logdir=logdir_train1)
  File "D:/proj_github/deep-voice-conversion/train1.py", line 60, in train
    launch_train_with_config(train_conf, trainer=trainer)
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorpack\train\interface.py", line 81, in launch_train_with_config
    model._build_graph_get_cost, model.get_optimizer)
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorpack\utils\argtools.py", line 181, in wrapper
    return func(*args, **kwargs)
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorpack\train\tower.py", line 173, in setup_graph
    train_callbacks = self._setup_graph(input, get_cost_fn, get_opt_fn)
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorpack\train\trainers.py", line 166, in _setup_graph
    self._make_get_grad_fn(input, get_cost_fn, get_opt_fn), get_opt_fn)
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorpack\graph_builder\training.py", line 232, in build
    all_grads = allreduce_grads(all_grads, average=self._average)  # #gpu x #param
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorpack\tfutils\scope_utils.py", line 84, in wrapper
    return func(*args, **kwargs)
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorpack\graph_builder\utils.py", line 140, in allreduce_grads
    summed = nccl.all_sum(grads)
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorflow\contrib\nccl\python\ops\nccl_ops.py", line 47, in all_sum
    return _apply_all_reduce('sum', tensors)
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorflow\contrib\nccl\python\ops\nccl_ops.py", line 228, in _apply_all_reduce
    shared_name=shared_name))
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorflow\contrib\nccl\ops\gen_nccl_ops.py", line 58, in nccl_all_reduce
    num_devices=num_devices, shared_name=shared_name, name=name)
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorflow\python\framework\op_def_library.py", line 787, in _apply_op_helper
    op_def=op_def)
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorflow\python\framework\ops.py", line 3414, in create_op
    op_def=op_def)
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorflow\python\framework\ops.py", line 1756, in __init__
    control_input_ops)
  File "C:\Users\mywind\AppData\Local\conda\conda\envs\python35\lib\site-packages\tensorflow\python\framework\ops.py", line 1592, in _create_c_op
    raise ValueError(str(e))
ValueError: Op type not registered 'NcclAllReduce' in binary running on mywind-PC. Make sure the Op and Kernel are registered in the binary running in this process. Note that if you are loading a saved graph which used ops from tf.contrib, accessing (e.g.) `tf.contrib.resampler` should be done before importing the graph, as contrib ops are lazily registered when the module is first accessed. while building NodeDef 'AllReduceGrads/NcclAllReduce'

Process finished with exit code 1

[juihsuanlee]

@wuzhiyu666 Hi, can you provide your code for me? I use the code on linux but it has some problem above. thank you so much!

[jiyuay]

@juihsuanlee I have made it and got a result. I use the code that put on the website(https://github.com/carlfm01/deep-voice-conversion). but made a little change in the code in convert.py hope it works to you

[juihsuanlee]

@wuzhiyu666 thanks for reply! Did you have ever had the error message below when you run the train1.py with arugment -case timit -gpu 0? TypeError: zip argument #1 must support iteration

[jiyuay]

sorry,I have not ever seen this error ,maybe you should ask the author ,I am a fresher, and it is my first time to use tensorflow,Sorry.

[juihsuanlee]

@wuzhiyu666 Hi, it's ok, thanks for your reply. And if you would not mind can you give me the pics of running state of the train1.py & train2.py ? I am trying to understand the error message. thank you so much!

[Huishou]

I'm running on Windows on a single GPU, you should migrate all the code that uses hparam.py, I changed all the code to use hparams.py, in most of the code you just have to change from default to Default, there is missing properties in Default and TrainX in hparams.py so, copy and paste the properties from hparam.py and replace the : for =

Nccl reduce may be caused by leaking wav files or the dataset path is incorrect, verify in the hparams.py, the other cause of ncclreduce is to use more than 1 GPU on windows.

My hparams.py, hope it helps. hparams.zip

How it works in Windows？ have changed the code of python2.7 to the code of Python 3.5. What should I do to run this program? I am a fresher, I hope you can help me run this program, thank you very much.

[Huishou]

@carlfm01 ，I used your code and read the comments, modified the code, but there are still some problems, I encountered the following error：

Process _Worker-5: Traceback (most recent call last): File "D:\Python set\lib\multiprocessing\process.py", line 252, in _bootstrap self.run() File "D:\Python set\lib\site-packages\tensorpack\dataflow\parallel.py", line 163, in run for dp in self.ds: File "D:\Python set\lib\site-packages\tensorpack\dataflow\common.py", line 116, in iter for data in self.ds: File "C:\Users\Administrator\PycharmProjects\deep-voice-conversion-windows\data_load.py", line 34, in get_data yield get_mfccs_and_phones(wav_file=wav_file) File "C:\Users\Administrator\PycharmProjects\deep-voice-conversion-windows\data_load.py", line 75, in get_mfccs_and_phones hp.Default.hop_length) File "C:\Users\Administrator\PycharmProjects\deep-voice-conversion-windows\data_load.py", line 152, in _get_mfcc_and_spec hp.Default.sr, hp.Default.n_fft, hp.Default.n_mels) # (n_mels, 1+n_fft//2) File "D:\Python set\lib\site-packages\librosa\filters.py", line 247, in mel lower = -ramps[i] / fdiff[i] ValueError: operands could not be broadcast together with shapes (1,569) (0,)

How can I fix it? I hope to get your reply.，Thank you！

[carlfm01]

Hi @Huishou, you can share the paths that you set in https://github.com/carlfm01/deep-voice-conversion/blob/master/params.py? Are you training on TIMIT or custom net1 dataset?

[Huishou]

Hi @carlfm01 , I run the train1 in Pycharm. I download the TIMIT for train1,and I fix the path :data_path = 'datasets/data/TIMIT/TRAIN///*.wav' , and it solve the problem which is IndexError: Cannot choose from an empty sequence",I think my path is right.
My env is win7 + tensorflow-gpu 1.5 + python3.5. And I download the ffmpeg, to slove the problem which is the warning RuntimeWarning: Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work warn("Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work", RuntimeWarning. But it now has this problem (ValueError: operands could not be broadcast together with shapes (1,569) (0,) ) , I do not know how to fix it. I am very grateful for your help and look forward to your reply!

[wishvivek]

@Huishou I faced the same broadcasting issue. I solved that by updating Librosa to version 0.6.2. Just do a pip install librosa (that installs 0.6.2). Should be okay after that.

[carlfm01]

@Huishou try updating tensorflow-gpu, you should use 1.8+, let me know if it solves the issue

[Huishou]

@wishvivek Thank you for your help. I solved that by using your method . And I will run the train1.py. It seem to work. But My GPU using is 5%, while the CPU using is above 85% ? How come this heppen ? My GPU is : device: 0, name: GeForce GTX 960M, pci bus id: 0000:01:00.0, compute capability: 5.0.

[Huishou]

@carlfm01 Thank you for your advise，I solve the broadcasting issue by updating Librosa. I run the train1.py. It has no Error now, but some WAN like :WRN [Deprecated]. And It seem to work,but very slow. Each epoch is above 5 minutes . like this: Epoch 4 (global_step 400) finished, time:5 minutes 32 seconds. My GPU using is 5%, while the CPU using is above 85% ,I think my GPU may not work. How long did it take when you run the train1.py ?

[Huishou]

@carlfm01 ,I try to run the train1.py, and it takes 3 hours to the Epoch 59 ..., and this is something wrong with it., and it does not work begin Epoch 60. The error is following: [0125 21:07:11 @base.py:272] Start Epoch 59 ... 0%| |0/100[00:00<?,?it/s]Process _Worker-2: Traceback (most recent call last): File "D:\Python set\lib\multiprocessing\process.py", line 252, in _bootstrap self.run() File "D:\Python set\lib\site-packages\tensorpack\dataflow\parallel.py", line 163, in run for dp in self.ds: File "D:\Python set\lib\site-packages\tensorpack\dataflow\common.py", line 116, in iter for data in self.ds: File "C:\Users\Administrator\PycharmProjects\deep-voice-conversion-windows\data_load.py", line 34, in get_data yield get_mfccs_and_phones(wav_file=wav_file) File "C:\Users\Administrator\PycharmProjects\deep-voice-conversion-windows\data_load.py", line 71, in get_mfccs_and_phones wav = read_wav(wav_file, sr=hp.Default.sr) File "C:\Users\Administrator\PycharmProjects\deep-voice-conversion-windows\audio.py", line 12, in readwav wav, = librosa.load(path, mono=mono, sr=sr, duration=duration) File "D:\Python set\lib\site-packages\librosa\core\audio.py", line 112, in load with audioread.audio_open(os.path.realpath(path)) as input_file: File "D:\Python set\lib\site-packages\audioread__init__.py", line 116, in audio_open raise NoBackendError() audioread.NoBackendError Process _Worker-1: Traceback (most recent call last): File "D:\Python set\lib\multiprocessing\process.py", line 252, in _bootstrap self.run() File "D:\Python set\lib\site-packages\tensorpack\dataflow\parallel.py", line 163, in run for dp in self.ds: File "D:\Python set\lib\site-packages\tensorpack\dataflow\common.py", line 116, in iter for data in self.ds: File "C:\Users\Administrator\PycharmProjects\deep-voice-conversion-windows\data_load.py", line 34, in get_data yield get_mfccs_and_phones(wav_file=wav_file) File "C:\Users\Administrator\PycharmProjects\deep-voice-conversion-windows\data_load.py", line 75, in get_mfccs_and_phones hp.Default.hop_length) File "C:\Users\Administrator\PycharmProjects\deep-voice-conversion-windows\data_load.py", line 153, in _get_mfcc_and_spec mel = np.dot(mel_basis, mag) # (n_mels, t) # mel spectrogram MemoryError

Have you encountered this problem at the time? How was it solved?

@wuzhiyu666 How can you solve the problem for the train1 ? I saw you have the similarly problem and solve it. I am very grateful for your help.

[carlfm01]

@carlfm01 Thank you for your advise，I solve the broadcasting issue by updating Librosa. I run the train1.py. It has no Error now, but some WAN like :WRN [Deprecated]. And It seem to work,but very slow. Each epoch is above 5 minutes . like this: Epoch 4 (global_step 400) finished, time:5 minutes 32 seconds. My GPU using is 5%, while the CPU using is above 85% ,I think my GPU may not work. How long did it take when you run the train1.py ?

The same happened to me, the net1 is not optimized for GPU

[carlfm01]

I see a memory error, maybe running out of memory? Never saw this error on a vm with 56GB RAM and a K80. Try decresing the batch size

[Huishou]

Hi，@carlfm01 .Are you running on the server? I think that the common computer does not have this configuration.My Notebook RAM is only 8G. The memory of my GPU is only 2G. Isn't my notebook running this program? Do I need to change my desktop? What configuration do I need for personal use? Is the i7 8700 and RTX2060 with 6G memory enough ? Besides，How long did it take when you run the train2.py ? I look forward to your reply！

[carlfm01]

I think is ok for this network, is not too big. Try with changing the batch size to 5 https://github.com/carlfm01/deep-voice-conversion/blob/5de50b955c8dd37d0948c0cc1bc965fa515aed88/params.py#L54

If you are planning to acquire one GPU try to go for one with 8GB.

[Huishou]

Hi，@carlfm01，Thank you for your reply and suggestions. I change the batch size to 5 for the train2.py. And the program did not report any error. But It run with a very slow speech. like this: [0127 09:03:06 @base.py:272] Start Epoch 1 ... 100%|##########|100/100[07:03<00:00, 0.24it/s] [0127 09:10:09 @base.py:282] Epoch 1 (global_step 100) finished, time:7 minutes 3 seconds. I think my GPU does not work ,either. How long did it take when you run the train2.py ?

[carlfm01]

Hi，@carlfm01，Thank you for your reply and suggestions. I change the batch size to 5 for the train2.py. And the program did not report any error. But It run with a very slow speech. like this: [0127 09:03:06 @base.py:272] Start Epoch 1 ... 100%|##########|100/100[07:03<00:00, 0.24it/s] [0127 09:10:09 @base.py:282] Epoch 1 (global_step 100) finished, time:7 minutes 3 seconds. I think my GPU does not work ,either. How long did it take when you run the train2.py ?

Two days on a k80, you can use nvidia smi to see if your gpu is being used https://unix.stackexchange.com/questions/38560/gpu-usage-monitoring-cuda

[Huishou]

Hi，@carlfm01，I can't understand the author's meaning now, and the intention to set up two networks. I want to know what role the net1 play? Is the TIMIT data set equivalent to multiple source speakers? The target sound of the author's net2 is the slt sound in arctic, which is a woman's voice. After running the convert.py, I listened to the bdl sound in arctic, which is a man's voice. So，Is this program just to convert the voice of a man in bdl into the voice of a woman in slt? What does the net1 work? I want to hear your opinion and look forward to your reply！

[carlfm01]

Hi, @Huishou TIMIT is a speech dataset aligned with its phonemes, the net1 is a speech recognizer trained with the speech and the phoenemes equivalent, then pass the recognized from net1 to net2, net2 is just synthetizing net1 prediction. That's why is important to use a dataset for net1 with lots of speakers to get a good prediction of what the speaker said.

[Sun-Ziyi]

      sorry ,I have not downloaded the timit data yet,I will download it  then tell you,Thank you very much!

hi，where I can download the timit data??? Can you give me a link??

[Sun-Ziyi]

hi, @carlfm01, I some questions about train1 1 you say: putting the timit data in the location of "data_path = 'Data/Train/.wav“， so the contents in Data/Train/ are all "wav"? 2 in net1, the training sets should be <wav, phone> pairs, so where phone (phoneme) should be put? 3 the timit data I found is like this: ( PHN \ txt \ wav \ WRD) so can you share the timit data you found? 4 the code carlfm01/deep-voice-conversion forked from andabi/deep-voice-conversion is excellent for me, a fresher; but can you write the implementation process step by step?

[Sun-Ziyi]

@carlfm01, after I put all the "wav" of the timit data into the path of 'Data/Train/*.wav', like this: I run train1, got: (results(1)(2)(3)(4) are sorted by the time order) (1) I dont understand what're meanings of the the sentences with the green? (2) in this part, what we saw are some training net parameters info? But why these training net parameters info can be given before the start of training (in my view, the trian1's goal is training the neural net parameters, so the these training net parameters info are just initial values?) ? (3) Start Epoch 1 ...came out, but appeared some problems: (4) in the end, IndexError: Cannot choose from an empty sequence came out: So, can you help me ???

[Sun-Ziyi]

@wuzhiyu666, hi , can you give some help? 1 can you share the timit data for me? 2 I dont understand that what datas should be choose and how to put datas in train1 \ eval1 \ train2 \ eval2, so, can you share those ( files form package, better) in baiduyun? thank you very much

[kay312]

@wuzhiyu666, hi , can you give some help? 1 can you share the timit data for me? 2 I dont understand that what datas should be choose and how to put datas in train1 \ eval1 \ train2 \ eval2, so, can you share those ( files form package, better) in baiduyun? thank you very much

兄弟，请问你搞定了吗？怎么解决掉呢？

[jiyuay]

啊，抱歉，年代太久远了，这些文件我都已经删掉了，timit数据集很好找的，还有哪些数据的问题你多试试就很快能搞定的。我记得当时最关键的问题是要在convert.py的那个位置把代码改成和图片里的一样。加油！

------------------ 原始邮件 ------------------ 发件人: "andabi/deep-voice-conversion" <notifications@github.com>; 发送时间: 2020年9月28日(星期一) 晚上7:13 收件人: "andabi/deep-voice-conversion"<deep-voice-conversion@noreply.github.com>; 抄送: "wuzhiyu"<308277419@qq.com>;"Mention"<mention@noreply.github.com>; 主题: Re: [andabi/deep-voice-conversion] I can't run it in window 10, could someone help me ? (#52)

@wuzhiyu666, hi , can you give some help?

1 can you share the timit data for me? 2 I dont understand that what datas should be choose and how to put datas in train1 \ eval1 \ train2 \ eval2, so, can you share those ( files form package, better) in baiduyun? thank you very much

兄弟，请问你搞定了吗？怎么解决掉呢？

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[kay312]

啊，抱歉，年代太久远了，这些文件我都已经删掉了，timit数据集很好找的，还有哪些数据的问题你多试试就很快能搞定的。我记得当时最关键的问题是要在convert.py的那个位置把代码改成和图片里的一样。加油！ … ------------------ 原始邮件 ------------------ 发件人: "andabi/deep-voice-conversion" <notifications@github.com>; 发送时间: 2020年9月28日(星期一) 晚上7:13 收件人: "andabi/deep-voice-conversion"<deep-voice-conversion@noreply.github.com>; 抄送: "wuzhiyu"<308277419@qq.com>;"Mention"<mention@noreply.github.com>; 主题: Re: [andabi/deep-voice-conversion] I can't run it in window 10, could someone help me ? (#52) @wuzhiyu666, hi , can you give some help? 1 can you share the timit data for me? 2 I dont understand that what datas should be choose and how to put datas in train1 \ eval1 \ train2 \ eval2, so, can you share those ( files form package, better) in baiduyun? thank you very much 兄弟，请问你搞定了吗？怎么解决掉呢？ — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

多谢了兄弟！

[kay312]

啊，抱歉，年代太久远了，这些文件我都已经删掉了，timit数据集很好找的，还有哪些数据的问题你多试试就很快能搞定的。我记得当时最关键的问题是要在convert.py的那个位置把代码改成和图片里的一样。加油！ … ------------------ 原始邮件 ------------------ 发件人: "andabi/deep-voice-conversion" <notifications@github.com>; 发送时间: 2020年9月28日(星期一) 晚上7:13 收件人: "andabi/deep-voice-conversion"<deep-voice-conversion@noreply.github.com>; 抄送: "wuzhiyu"<308277419@qq.com>;"Mention"<mention@noreply.github.com>; 主题: Re: [andabi/deep-voice-conversion] I can't run it in window 10, could someone help me ? (#52) @wuzhiyu666, hi , can you give some help? 1 can you share the timit data for me? 2 I dont understand that what datas should be choose and how to put datas in train1 \ eval1 \ train2 \ eval2, so, can you share those ( files form package, better) in baiduyun? thank you very much 兄弟，请问你搞定了吗？怎么解决掉呢？ — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

兄弟，我现在跑通了，但是生成的音频文件都是一秒的，请问你知道怎么解决吗？

[jiyuay]

我记得有batchsize等等参数，你可能要试试那些。

---Original--- From: "Kay-GH"<notifications@github.com> Date: Tue, Sep 29, 2020 16:54 PM To: "andabi/deep-voice-conversion"<deep-voice-conversion@noreply.github.com>; Cc: "Mention"<mention@noreply.github.com>;"jiyu_ay"<308277419@qq.com>; Subject: Re: [andabi/deep-voice-conversion] I can't run it in window 10, could someone help me ? (#52)

啊，抱歉，年代太久远了，这些文件我都已经删掉了，timit数据集很好找的，还有哪些数据的问题你多试试就很快能搞定的。我记得当时最关键的问题是要在convert.py的那个位置把代码改成和图片里的一样。加油！ … ------------------ 原始邮件 ------------------ 发件人: "andabi/deep-voice-conversion" <notifications@github.com>; 发送时间: 2020年9月28日(星期一) 晚上7:13 收件人: "andabi/deep-voice-conversion"<deep-voice-conversion@noreply.github.com>; 抄送: "wuzhiyu"<308277419@qq.com>;"Mention"<mention@noreply.github.com>; 主题: Re: [andabi/deep-voice-conversion] I can't run it in window 10, could someone help me ? (#52) @wuzhiyu666, hi , can you give some help? 1 can you share the timit data for me? 2 I dont understand that what datas should be choose and how to put datas in train1 \ eval1 \ train2 \ eval2, so, can you share those ( files form package, better) in baiduyun? thank you very much 兄弟，请问你搞定了吗？怎么解决掉呢？ — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

兄弟，我现在跑通了，但是生成的音频文件都是一秒的，请问你知道怎么解决吗？

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[kay312]

我记得有batchsize等等参数，你可能要试试那些。 … ---Original--- From: "Kay-GH"<notifications@github.com> Date: Tue, Sep 29, 2020 16:54 PM To: "andabi/deep-voice-conversion"<deep-voice-conversion@noreply.github.com>; Cc: "Mention"<mention@noreply.github.com>;"jiyu_ay"<308277419@qq.com>; Subject: Re: [andabi/deep-voice-conversion] I can't run it in window 10, could someone help me ? (#52) 啊，抱歉，年代太久远了，这些文件我都已经删掉了，timit数据集很好找的，还有哪些数据的问题你多试试就很快能搞定的。我记得当时最关键的问题是要在convert.py的那个位置把代码改成和图片里的一样。加油！ … ------------------ 原始邮件 ------------------ 发件人: "andabi/deep-voice-conversion" <notifications@github.com>; 发送时间: 2020年9月28日(星期一) 晚上7:13 收件人: "andabi/deep-voice-conversion"<deep-voice-conversion@noreply.github.com>; 抄送: "wuzhiyu"<308277419@qq.com>;"Mention"<mention@noreply.github.com>; 主题: Re: [andabi/deep-voice-conversion] I can't run it in window 10, could someone help me ? (#52) @wuzhiyu666, hi , can you give some help? 1 can you share the timit data for me? 2 I dont understand that what datas should be choose and how to put datas in train1 \ eval1 \ train2 \ eval2, so, can you share those ( files form package, better) in baiduyun? thank you very much 兄弟，请问你搞定了吗？怎么解决掉呢？ — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe. 兄弟，我现在跑通了，但是生成的音频文件都是一秒的，请问你知道怎么解决吗？ — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

多的，多谢多谢