Maximum likelihood training and adaptation of embedded speech recognizers for mobile environments

Youngkyu Cho, Dongsuk Yook

    Research output: Contribution to journalArticlepeer-review

    3 Citations (Scopus)

    Abstract

    For the acoustic models of embedded speech recognition systems, hidden Markov models (HMMs) are usually quantized and the original full space distributions are represented by combinations of a few quantized distribution prototypes. We propose a maximum likelihood objective function to train the quantized distribution prototypes. The experimental results show that the new training algorithm and the link structure adaptation scheme for the quantized HMMs reduce the word recognition error rate by 20.0%.

    Original languageEnglish
    Pages (from-to)160-162
    Number of pages3
    JournalETRI Journal
    Volume32
    Issue number1
    DOIs
    Publication statusPublished - 2010 Feb

    Keywords

    • Embedded speech recognition
    • Maximum likelihood distribution clustering (MLDC)
    • Quantized HMM

    ASJC Scopus subject areas

    • Electronic, Optical and Magnetic Materials
    • General Computer Science
    • Electrical and Electronic Engineering

    Fingerprint

    Dive into the research topics of 'Maximum likelihood training and adaptation of embedded speech recognizers for mobile environments'. Together they form a unique fingerprint.

    Cite this