Idiap on LinkedIn Idiap youtube channel Idiap on Twitter Idiap on Facebook
Personal tools
You are here: Home Research Resources ISS

ISS

— filed under:

The Idiap Speech Scripts (ISS) is a collection of speech databases and dictionaries, and for training and testing of models for ASR. The scripts in turn are reliant on many other packages including HTK/HTS, Juicer and the ICSI speech tools.

Features include:

  • GMM (HTK/HTS) and MLP (QuickNet) training in the same framework
  • Adaptation and speaker adaptive training
  • Tandem
  • Testing with hvite, hdecode and juicer
  • Plays well with SGE (no knowledge of SGE required)
  • Runs seamlessly either on the grid or interactively
  • WFST composition with AT&T or OpenFST toolkits
  • Large and growing collection of phone sets
  • Automatic decision tree question sets for many languages
  • Automatic decision tree threshold tuning using MDL or state count
  • Features from HTK, SPTK or tracter (others trivially)
  • HTK extended filenames, or not, your choice
  • Database maintenance tools
  • Out of vocabulary pronunciations via phonetisaurus


All the packages are on GitHub:

Core scripts:
https://github.com/idiap/iss

Dictionary maintenance
https://github.com/idiap/iss-dicts

Exemplars for the Wall Street Journal and Aurora-4 tasks
https://github.com/idiap/iss-wsj
https://github.com/idiap/iss-aurora4

Some other related speech processing packages are also available:

Juicer, a WFST based decoder for ASR
https://github.com/idiap/juicer

Tracter, a data-flow framework
https://github.com/idiap/tracter

SSP, a signal processing package written in python
https://github.com/idiap/ssp

 

Most releases are under a BSD licence, which is quite permissive.

Document Actions
Resource Information
Resource type: software
URL: https://github.com/idiap/iss
Date: Aug 13, 2013
Lifespan: Active
Audience: Speech & Audio researchers
Access: Web
Ownership: Idiap Research Institute
Distribution: Web & GitHub
License:
Contact: Philip GARNER
+41 277 217 768