zhenzi-weng / deepsc-s Goto Github PK
View Code? Open in Web Editor NEWSemantic Communication Systems for Speech Transmission
License: MIT License
Semantic Communication Systems for Speech Transmission
License: MIT License
hello,
i have run your code, but the loss is hard to get 5e-7, specifically,It converges when the loss reaches 5e-6. Is there such a big difference between different data sets?
Dear author, I would like to get your Testing Stage code file as described in your paper.
Best wishes!
Dear author, when I run your code, it appears this errors as in figure 1 below.
So, how can I fixe it? In figure 2 is folder for this code. The "path to save the trained datatrain" folder does not have any created file after running main.py file.
Many thanks.
***************** start train *****************
WARNING:tensorflow:_input
is not a valid node name. Accepted names conform to Regex /re.compile('^[A-Za-z0-9.][A-Za-z0-9_.\\/>-]$')/
WARNING:tensorflow:_input
is not a valid node name. Accepted names conform to Regex /re.compile('^[A-Za-z0-9.][A-Za-z0-9_.\\/>-]$')/
The version I am using is TensorFlow2.4.1, but the first epoch of training shows a loss of nan
I downloaded noise_trainset_28spk_wav from Edinburgh DataShare as a dataset, and changed the sampling rate in the code to sr=48kHz according to the information given by the website, and generated the corresponding .tfrecords file.
When I run main.py now, I'm told UnicodeDecodeError: 'utf-8' codec can't decode byte 0xbe in position 77: invalid start byte, and can't find the cause of the problem so far.
why the channel output(8128128) is larger than the source size(128*128)? It is unfair for the tradition baseline!
I use the clean_trainset_28spk_wav.zip (2.315Gb)
take error "Sampling rate is expected to be 8kHz!"
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.