GithubHelp home page GithubHelp logo

fgnt / paderbox Goto Github PK

View Code? Open in Web Editor NEW
36.0 36.0 127.0 1.97 MB

Paderbox: A collection of utilities for audio / speech processing

Home Page: https://pypi.org/project/paderbox

License: MIT License

Makefile 0.02% Shell 0.47% Python 98.70% Cython 0.81%
audio speech toolbox

paderbox's People

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

paderbox's Issues

Citation with Zenodo

heyo!

I saw in your README that you wrote:

There is no clear way how to cite this repository for research. However, we would be grateful if directly import from this repository if you use, e.g., the STFT. We are also fine when you copy the code as long as it remains visible where you copied the code from.

And just wanted to suggest that you add the repository to Zenodo, which would give a clear way to cite it. Happy Holidays!

Question about Amplitude Modulation Spectrogram

Hi, folks.

I'm researching about modulation spectrogram. I've read a bunch of articles and past implementations in Matlab, then I'm trying to code one right now in Python.

I found this repository of yours, as well as an IEEE article from your lab describing it. The article is really helpful.

I would like to know if there's any reason about removing it from your repository. It was added in this commit, but then removed and declared obsolete in this one.

Thanks in advance :)

Some questions

Thanks a lot for sharing this, that looks really nice ! :-)

  • transform_to_baseband makes complete sense. Question : did you use it as input features to DNN, does it make any difference with or without the phase corrections? I intended to try but didn't do it yet.

  • About the approach for the STFT / iSTFT, in particular about the biorthogonal synthesis window.

    • I don't find the dissertation from which the equations are derived. Would it be possible to have a link to it? Or some other resources that explains the approach? It amounts to divide the time-domain synthesized signal by the sum of the overlap-added squared analysis windows right?
    • If my understanding is correct, iSTFT(STFT(x)) == x for any window and any shift. However, if the STFT is modified (which is usually the case if we are about to take the iSTFT), the use of the biorthogonal synthesis window will bring time domain amplitude modulation. Did you make any experiments to quantify those artifacts and compare them against using the constant scaling factor? Especially for bad window choices?

Cheers,

cannot find lazy_dataset.database

when I try running the file pb_bss/testing/dummy_data.py ๏ผŒ it errors that No module named 'lazy_dataset'.

I Follow all the command mentioned in readme. But it still doesn't work. Is there any other libraries should be intalled via pip?

By the way, what does (:Could not build wheels for paderbox, which is required to install pyproject.toml-based projects
) mean?

nested.get_by_path fails with default

@thequilo Do you remember, why the default is only considered for KeyError and IndexError?
When should get_by_path fail?

except Exception as e:
# Indexing a custom type can raise any exception, in which case
# we try to broadcast
# Not sure if broadcasting makes sense for lists/tuples. It is
# hard to check for custom sequences because of str, so
# sequences are broadcasted here
if allow_early_stopping and not isinstance(container, Mapping):
return container
# We can't add another more specific except block because we have to
# catch all exceptions for the broadcasting. Assuming here that
# custom containers raise KeyErrors and IndexErrors correctly when
# the indexed element is not found
if isinstance(e, (KeyError, IndexError)):
if default is not _NO_VALUE:
return default
raise

Here, an extension of the doctest. IMHO the last example should work.

>>> d = {'a': 'b', 'c': {'d': {'e': 'f'}, 'g': [1, [2, 3], 4]}}
...
>>> get_by_path(d, 'c.b.c', default=42)
42
>>> get_by_path(d, 'a.b.c', default=42)
Traceback (most recent call last):
...
TypeError: string indices must be integers

The leafs in a nested structure can have any type and hence any Exception could be thrown.
So I don't get, why there is an if statement.

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

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.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.