lixikun
lixikun
me too, what's does this mean?
thank you for your good job!!
hi,you just modify the imports in file helper.py , the path is seq2seq/seq2seq/contrib/seq2seq/helper.py. I has tested in tensorflow 1.4,and the test result is ok! from tensorflow.python.ops.distributions import bernoulli from tensorflow.python.ops.distributions...
how about this issue? someone can solve it ?
There is speed parameter in the model.inference interface, the following is the explaination. ` speed: The speed rate of the generated audio. Defaults to 1.0. (can produce artifacts if far...
Do we have a solution for this Exception? waveIn_RecordingStopped: exception=NAudio.MmException: WaveStillPlaying calling waveUnprepareHeader в NAudio.MmException.Try(MmResult result, String function) в NAudio.Wave.WaveInBuffer.Reuse()
meet the same question, someone konw this?
> To be solved with a V5 release Great job!
先看下你的采样率是否是16K的，采样率一样基本不会有太大的差别
这个去掉 跳过 refine text 的步骤，听起来还可以。感觉有些字它不认识。。