danijel3 / clarinstudiokaldi Goto Github PK
View Code? Open in Web Editor NEWA baseline Automatic Speech Recognition system for Polish based on Kaldi.
License: Apache License 2.0
A baseline Automatic Speech Recognition system for Polish based on Kaldi.
License: Apache License 2.0
In the paper it is said that
The corpus consists of 317 speakers recorded in 554 sessions
and
56 random sessions were chosen to be in the test set, and the remaining 499 session were stored in the test [mistake, it is train] set.
Does it mean that the overfitting in terms of speakers was possible? I mean the same speaker can be in train and test set. The results don't seem reliable...
Is there any information about speaker ID in database? Then it could be possible to test the baseline system truly
Hello @danijel3 ! Thanks for the great project!
I want to try to perform speech recognition for polish. How can I get transcriptions having only audio recordings of speech?
Thanks!
phonetisaurus_apply, that is used in clarin_prepare_dict.sh, was renamed into phonetisaurus-apply on 29.08.2017
Which version of Kaldi should currently be used with ClarinStudio setup?
Is it still the one "pulled from the github on Aug, 11. 2017"?
I think that some of the scripts can run without SRILM tool. However, in local_clarin/clarin_pl_data_prep.sh there is a ngram command, that comes from SRILM, as far as I understand (but it does not perform any advanced task)
The corpus is said to be created for commercial and non-commercial use. Maybe it would be nice to allow the scripts to run with only open-licence tools too? The SRILM in this place is probably the only one limitation.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ 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. ๐๐๐
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 โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.