Unsupervised Word Segmentation using K Nearest Neighbors

Tzeviya Sylvia Fuchs, Yedid Hoshen, Joseph Keshet

Research output: Contribution to journalConference articlepeer-review

2 Scopus citations

Abstract

In this paper, we propose an unsupervised kNN-based approach for word segmentation in speech utterances. Our method relies on self-supervised pre-trained speech representations, and compares each audio segment of a given utterance to its k nearest neighbors within the training set. Our main assumption is that a segment containing more than one word would occur less often than a segment containing a single word. Our method does not require phoneme discovery and is able to operate directly on pre-trained audio representations. This is in contrast to current methods that use a two-stage approach; first detecting the phonemes in the utterance and then detecting word-boundaries according to statistics calculated on phoneme patterns. Experiments on two datasets demonstrate improved results over previous single-stage methods and competitive results on state-of-the-art two-stage methods.

Original languageAmerican English
Pages (from-to)4646-4650
Number of pages5
JournalProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
Volume2022-September
DOIs
StatePublished - 2022
Event23rd Annual Conference of the International Speech Communication Association, INTERSPEECH 2022 - Incheon, Korea, Republic of
Duration: 18 Sep 202222 Sep 2022

Bibliographical note

Publisher Copyright:
Copyright © 2022 ISCA.

Keywords

  • Unsupervised speech processing
  • language acquisition
  • unsupervised clustering
  • unsupervised segmentation

Fingerprint

Dive into the research topics of 'Unsupervised Word Segmentation using K Nearest Neighbors'. Together they form a unique fingerprint.

Cite this