CNN-based Spoken Term Detection and Localization without Dynamic Programming

TL;DR

This paper introduces a CNN-based spoken term detection method that predicts and localizes terms in speech without dynamic programming, leveraging existing embeddings for efficient and simultaneous detection of in-vocabulary and out-of-vocabulary terms.

Contribution

The proposed algorithm detects and localizes spoken terms directly from speech signals using CNNs and existing embeddings, eliminating the need for task-specific embedding training and dynamic programming.

Findings

Effective detection of in-vocabulary and out-of-vocabulary terms

Simultaneous localization without dynamic programming

Evaluated on read speech corpora with promising results

Abstract

In this paper, we propose a spoken term detection algorithm for simultaneous prediction and localization of in-vocabulary and out-of-vocabulary terms within an audio segment. The proposed algorithm infers whether a term was uttered within a given speech signal or not by predicting the word embeddings of various parts of the speech signal and comparing them to the word embedding of the desired term. The algorithm utilizes an existing embedding space for this task and does not need to train a task-specific embedding space. At inference the algorithm simultaneously predicts all possible locations of the target term and does not need dynamic programming for optimal search. We evaluate our system on several spoken term detection tasks on read speech corpora.