Comments (9)
@enhuiz thanks for another one feedback. I agree with you that it should be exponents = 1e-4 ** exponents
! However, I think for model it is still interpretable at some point since we cat sin
and cos
, which makes more or less individual (locally) encoding for noise levels at range of [0, 1]. But, obviously, what happens currently, it is not good. Please, report on your experiment with the second approach! I will also try.
from wavegrad.
Hi @ivanvovk, I have tried to fix the positional encoding and retrain the model, though the grad_norm get lower, the test result seems much worse (from both the test loss curve and the generated samples).
I guess there could be some other issues, so I check the other part of the code and find here is a mismatch between the implementation and the paper (formula 11).
WaveGrad/model/diffusion_process.py
Line 103 in d230621
An sqrt() seems lost here. I'll fix it and try again.
from wavegrad.
@enhuiz yeah, I fixed PE and got the same problems. And yeah, you're right, sqrt() is missed here, need to fix it also.
from wavegrad.
@enhuiz seems like for me sqrt() update solves the problem and now test samples look good. How it does for you?
from wavegrad.
The loss curve looks better than the previous one, l1_spec_test_batch_loss and total loss are lower, l1_test_batch_loss is higher which is acceptable as it is measure on the audio wave. Training grad and total loss are both lower.
I think I'm still in the early stage. I use niters=1000 to train and niters=50 to test, the audio quality of the fixed version seems not significantly better than the previous one.
from wavegrad.
WaveGrad/model/diffusion_process.py
Line 116 in d230621
I find changing this t to t+1 helps remove the noise in the generated sample after fix pe and sqrt, you may check the following samples:
I guess here we need the current sqrt cumprod instead of the previous one.
from wavegrad.
@enhuiz yes, I agree! For me this change improved the quality even more! Somehow missed it when made the implementation... Thanks for revealing all these bugs, man, I really appreciate it. Now it seems to work fine.
from wavegrad.
@ivanvovk Good to know it, you are welcome!
from wavegrad.
Closing issue since it is solved.
from wavegrad.
Related Issues (20)
- ValueError: low >= high HOT 2
- inference.py seems not loading the specified checkpoint HOT 1
- Were your `generated_samples` generated using a model trained with AMP? HOT 2
- predict_start_from_noise HOT 2
- best noise schedule HOT 1
- schedules model for other dataset and different sample rate HOT 4
- TTS without Text? HOT 2
- Training so slow HOT 3
- How to make it work with TPU? HOT 2
- Audio quality improvements HOT 6
- The order of upsampling_dilations HOT 1
- Interpolation and Conv order in Upsample module HOT 1
- slow training in single GPU HOT 1
- Using NVIDIA RTX 3090 GPU?
- Static Noise with f_max = 10000
- Poor Synthesis Quality on 44k Sample Rate HOT 1
- Evaluation tools
- Unable to load the pre-trained parameters for inference HOT 2
- Matplotlib API change & NaNs for short clips & new hop_length HOT 27
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from wavegrad.