Abstract:Keyword search (KWS) is an important application of spoken language technology. The technique of Large Vocabulary Continuous Speech Recognition (LVCSR) is playing an important role in KWS system. However, for a language with large vocabulary and relatively insufficient text corpus, the vocabulary size keeps going up very quickly with the increasing amount of text, as we observed in Tamil. This brings difficulty in training a reliable language model, which may undermine KWS performance. Subword unit has been successfully employed in KWS system to handle out-of-vocabulary (OOV) problem. Inspired by this, we propose a novel subword scheme from the perspective of pronunciation to alleviate the large vocabulary problem. We find that the subword-based system outperforms our best word-based system on Tamil conversational telephone speech. The experiment of system combination shows that, over the best word-based system, a single subword-based system contains more complementary information than the total of that of the other three word-based systems.

Research on continuous Chinese speech recognition system and knowledge based search strategies

ON CHINESE CONTINUOUS SPEECH RECOGNITION SYSTEM AND KNOWLEDGE BASED SEARCH STRATEGIES SONG Zhanjiang

Improving the Syllable-Synchronous Network Search Algorithm for Word Decoding in Continuous Chinese Speech Recognition

A two-layer lexical tree based beam search in continuous Chinese speech recognition

Research on Inter-Syllable Context-Dependent Acoustic Unit for Mandarin Continuous Speech Recognition.

Algorithm for Mandarin Continuous Speech Recognition Based on Context-Dependent Unit Between Syllables

Context Dependent Syllable Acoustic Model For Continuous Chinese Speech Recognition

A Novel Beam Search Algorithm Of Speech Recognition For Voice Command Control

Keyword Spotting Based on Syllable Confusion Network.

Two-level Search Strategy Based on Reverse HMM and Its Applications

Subword scheme for keyword search

Fast Fuzzy Keyword Spotting Using Syllable Confusion Network Indexing

STUDY ON FRAMEWORK FOR CHINESE PRONUNCIATION VARIATION MODELING

Continuous speech recognition method and continuous speech recognition system

AN AUTOMATIC SPEECH RECOGNITION STRATEGY DIRECTED BY THE SEMANTIC KNOWLEDGE IN DIALOGUE SYSTEM

Improved context-dependent acoustic modeling for continuous Chinese speech recognition

A Fast Fuzzy Keyword Spotting Algorithm Based on Syllable Confusion Network.

Research on Optimal Algorithm of Likelihood Computation in Continuous Speech Recognition

A Two-Step Keyword Spotting Method Using Fuzzy Search Algorithm

Toward On-Line Learning of Chinese Continuous Speech Recognition System.

Cross-word Search Algorithm Based on Two-Layer Lexical Tree for Speech Recognition