The Influence of Dataset Partitioning on Dysfluency Detection Systems

  • This paper empirically investigates the influence of different data splits and splitting strategies on the performance of dysfluency detection systems. For this, we perform experiments using wav2vec 2.0 models with a classification head as well as support vector machines (SVM) in conjunction with the features extracted from the wav2vec 2.0 model to detect dysfluencies. We train and evaluate the systems with different non-speaker-exclusive and speaker-exclusive splits of the Stuttering Events in Podcasts (SEP-28k) dataset to shed some light on the variability of results w.r.t. to the partition method used. Furthermore, we show that the SEP-28k dataset is dominated by only a few speakers, making it difficult to evaluate. To remedy this problem, we created SEP-28k-Extended (SEP-28k-E), containing semi-automatically generated speaker and gender information for the SEP-28k corpus, and suggest different data splits, each useful for evaluating other aspects of methods for dysfluency detection.

Export metadata

Additional Services

Search Google Scholar
Metadaten
Author:Sebastian P. BayerlORCiD, Dominik Wagner, Elmar NöthORCiD, Tobias BockletORCiD, Korbinian RiedhammerORCiD
DOI:https://doi.org/10.1007/978-3-031-16270-1_35
ISBN:9783031162695
ISSN:0302-9743
Parent Title (English):Text, Speech, and Dialogue
Publisher:Springer International Publishing
Place of publication:Cham
Document Type:Part of a Book
Language:English
Date of first Publication:2022/09/16
Reviewed:Begutachtet/Reviewed
Release Date:2024/08/01
Tag:Computer Science; Informatics
Pagenumber:14
First Page:423
Last Page:436
Konferenzangabe:25th International Conference, TSD 2022, Brno, Czech Republic, September 6–9, 2022
institutes:Fakultät Informatik
Zentrum für Künstliche Intelligenz (KIZ)
Research Themes:Digitalisierung & Künstliche Intelligenz
Verstanden ✔
Diese Webseite verwendet technisch erforderliche Session-Cookies. Durch die weitere Nutzung der Webseite stimmen Sie diesem zu. Unsere Datenschutzerklärung finden Sie hier.