UNTREF Speech Workshop
From Robert-Depot
Contents
Introduction
A short 1-2 day workshop introduction open source speech recognition and speech synthesis technologies for making interactive artworks. We use pre-compiled open-source tools (CMU Sphinx ASR and Festival TTS), and focus on the importance of language model content and interaction context as context context for creating meaning in the work.
Background
- If Things Can Talk, What Do They Say? If We Can Talk to Things, What Do We Say? Natalie Jeremijenko. 2005-03-05 [1]
- also see the responses by Simon Penny, Lucy Suchmann, and Natalie.
- Dialogue with a Monologue: Voice Chips and the Products of Abstract Speech. [2]
Speech Recognition
Introduction
- pocketsphinx on win32 - http://www.aiaioo.com/cms/index.php?id=28
Installing CMU Sphinx
http://cmusphinx.sourceforge.net/wiki/download/
Language Models
Acoustic models versus language models.
Grammars versus Satistical Language Models.
Using sphinx
- open a terminal. Windows, Run->Cmd.
- change to the pocketsphinx directory.
cd Desktop\untref_speech\pocketsphinx-0.8-win32\bin\Release
- run the pocketsphinx command:
pocketsphinx_continuous.exe -hmm ..\..\model\hmm\en_US\hub4wsj_sc_8k -dict ..\..\model\lm\en_US\cmu07a.dic -lm ..\..\model\lm\en_US\hub4.5000.DMP- this should transcribe live from the microphone.
Training your own Models
grammer is trivial.
slm, can use online tools. or try the sphinxtrain packages.
Programming with Speech Recognition
Processing. Sphinx4, the java interface.
Python or c++, command line, android. pocketsphinx.
Speech Synthesis
Introduction
FestVox. CMU Speech group.
Festival from University of Edinburgh.
