Abstract:<p>Keyword search (KWS) means searching for keywords given by the user from continuous speech. Conventional KWS systems are based on Automatic Speech Recognition (ASR), where the input speech has to be first processed by the ASR system before keyword searching. In the recent decade, as deep learning and deep neural networks (DNN) become increasingly popular, KWS systems can also be trained in an end-to-end (E2E) manner. The main advantage of E2E KWS is that there is no need for speech recognition, which makes the training and searching procedure much more straightforward than the traditional ones. This article proposes an E2E KWS model, which consists of four parts: speech encoder-decoder, query encoder-decoder, attention mechanism, and energy scorer. Firstly, the proposed model outperforms the baseline model. Secondly, we find that under various supervision, character or phoneme sequences, speech or query encoders can extract the corresponding information, resulting in different performances. Moreover, we introduce an attention mechanism and invent a novel energy scorer, where the former can help locate keywords. The latter can make final decisions by considering speech embeddings, query embeddings, and attention weights in parallel. We evaluate our model on low resource conditions with about 10-hour training data for four different languages. The experiment results prove that the proposed model can work well on low resource conditions.</p>

Efficient Keyword Spotting System for Information Retrieval

Two-stage Keyword Spotting System Based on Syllable Graphs

A Two-Step Keyword Spotting Method Using Fuzzy Search Algorithm

Fast Fuzzy Keyword Spotting Using Syllable Confusion Network Indexing

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

Keyword Spotting Based on Syllable Confusion Network.

A New Keyword Spotting Approach for Spontaneous Mandarin Speech

Keyword Spotting Based on Phoneme Confusion Matrix

Keyword-specific normalization based keyword spotting for spontaneous speech

Word Spotting Based on a Posterior Measure of Keyword Confidence

A Keyword Spotting Method

Audio-visual Keyword Spotting Based on Adaptive Decision Fusion under Noisy Conditions for Human-Robot Interaction.

A VOCABULARY-INDEPENDENT KEYWORD SPOTTER FOR SPONTANEOUS CHINESE SPEECH

LEXICAL ACCESS-BASED CONFIDENCE MEASURE FOR A SPANISH KEYWORD SPOTTING SYSTEM

SLiCK: Exploiting Subsequences for Length-Constrained Keyword Spotting

Audio-visual Keyword Spotting for Mandarin Based on Discriminative Local Spatial-Temporal Descriptors.

TDT-KWS: Fast And Accurate Keyword Spotting Using Token-and-duration Transducer

Syllable Based Audio Search Using Confusion Network Arc As Indexing Unit

Efficient graph search algorithm for embedded speech recognition

Improved Keyword Spotting System by Optimizing Posterior Confidence Measure Vector Using Feed-Forward Neural Network.

End-to-end keyword search system based on attention mechanism and energy scorer for low resource languages