-
This is what I get on step 2. I tried working with the fixes that people suggested on the other posts, but have still not had any luck. Any suggestions would be greatly appreciated!
ValueErrorTra…
-
I use Anaconda, on Windows 7 with conda activate.
Tensorflow 1.5
librosa 0.5
2 GPU GTX 1060 6GB
Training network with train.py --data_dir=corpus --num_steps=16000
after few second i got this e…
-
URL - https://mxnet.incubator.apache.org/tutorials/tensorrt/inference_with_trt.html
Broken Links
─ https://mxnet.incubator.apache.org/versions/master/tutorials/tensorrt/_static/tutorials/tensorrt/w…
-
Installed all requirements.txt in conda enviroment but modules "train" and "wavenet_vocoder" give errors.
cannot import name 'linear_quantize' from 'wavenet_vocoder.util'
cannot import name 'buil…
-
-
Hey, when i try to synthesis wav from mel files , this error occurs:
python synthesis.py --conditional=./output_mel/0_mel.npy ./wavenet_premodel/20180510_mixture_lj_checkpoint_step000320000_ema.pth…
-
Hi,
When run inference.py for a test set which contains 500 mel files, the GPU memory was increasing until the OOM assert. Was the GPU memory freed automatically, or by some API?
assert info:
G…
-
The default mu_quantization value is 256, and I think higher value of mu_quantization may produce better quality audio.
I did these modifications to test mu_quantization value of 512:
1. modify va…
nanma updated
6 years ago
-
Is there a reference implementation of a end-to-end TTS based on nv-wavenet which can be deployed on a NVIDIA Jetson TX2 GPU to get synthesis done within a second or two? Any input to deploy a wavenet…
-
## 一言でいうと
WaveNetは精度は良いものの、生成は逐次的(過去の自身の生成結果を利用する)ためとても生成に時間がかかるという問題があった。そこで、IAF(Inverse Autoregressive Flow)という再帰的な実行で分布近似を行うようなモデルを利用し、(自身の生成結果でなく)ノイズから徐々にあるべき音の分布へと近づけていき、最終的に訓練されたWaveNetの分布と近く…