tips:speech:simon [Galileo Labs]

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== simon ======

speech model basically consists of two parts:
  * Language model: Describes all existing words and what sentences are grammatically correct. It will be created from your active scenario
  * Acoustic model: Describes how words sound. It will be either built solely through your voice recordings (training) or with the help of a base model


SPHINX backend" uses:
  * PocketSphinx recognizer
  * SphinxTrain model generation backend

models:
  * US English HUB4 WSJ SPHINX model. This model is part of pocketsphinx-extra where it is called \"hub4_wsj_sc_3s_8k.cd_semi_5000\".
===== install sphinx =====

prereq
<code bash>
sudo apt install bison gradle openjdk-8-jdk
</code>

build from sources http://cmusphinx.sourceforge.net/wiki/download, not use ubuntu packages




===== install =====




<code bash>
sudo apt install simon

sudo apt install libsphinxbase3 python-sphinxbase sphinxbase-utils pocketsphinx sphinxtrain
</code>

edit 
<file ini /usr/lib/sphinxtrain/etc/sphinx_train.cfg> 
$CFG_CD_TRAIN = 'no';
</file>

===== compile =====

https://phabricator.kde.org/source/simon/

<code bash>
sudo apt-get build-dep simon

git clone git://anongit.kde.org/simon
cd simon
export QT_VERSION=qt4
rm -fR build
./build_ubuntu.sh
</code>


==== model ====

=== adapting ===

http://cmusphinx.sourceforge.net/wiki/tutorialadapt

<code bash>
sudo apt install pocketsphinx-en-us/xenial

</code>

<code bash>
mkdir /tmp/model
cd /tmp/model
cp -a /usr/share/pocketsphinx/model/en-us/en-us .
cp /usr/share/pocketsphinx/model/en-us/cmudict-en-us.dict .
cp /usr/share/pocketsphinx/model/en-us/en-us.lm.bin .
</code>




=== training ===


training from http://cmusphinx.sourceforge.net/wiki/tutorialamdown

* download [[model|http://www.speech.cs.cmu.edu/databases/an4/an4_sphere.tar.gz]]
* enter and sphinxtrain -t an4 setup




==== old ====

For small amount of data for training without context-dependent models use continuous model instead of semi-continuous one and this error will be solved.

This warning is caused by the fact you disabled context-dependent training and used semi-continuous model. You can ignore this error until you have more data and use context-dependent training.


training

<code bash>
mkdir /tmp/test
cd /tmp/test

sphinxtrain -t test setup
</code>

now we have
<code bash>
.
└── etc
│   ├── feat.params
│   ├── sphinx_train.cfg
│   ├── test.dic
│   └── test.filler

</code>

adjust sphinx_train.cfg TODO