Show Less
Restricted access

Nordic Prosody

Proceedings of the XIIth Conference, Trondheim 2016

Edited By Jardar Eggesbö Abrahamsen, Jacques Koreman and Wim van Dommelen

This volume contains articles based on the presentations given at the Nordic Prosody XII conference, which was held at the Norwegian University of Science and Technology (Trondheim, Norway) in August 2016. The contributors investigate various prosodic aspects, including intonation, rhythm, speaking rate, intensity, and breathing, using approaches ranging from phonetic and phonological analysis to speech technology methods. While most of the studies examine read speech, some of them explore the prosodics of spontaneous speech. The languages that receive most attention are Norwegian, Swedish and Icelandic as well as Estonian, Latgalian and Polish. In addition to the larger Nordic languages, several papers focus on regional languages spoken in these areas.

Show Summary Details
Restricted access

Capturing respiratory sounds with throat microphones (Marcin Włodarczak / Mattias Heldner)


Marcin Włodarczak, Mattias Heldner

Capturing respiratory sounds with throat microphones

Abstract: This paper presents results of a pilot study using throat microphones for recording respiratory sounds. We demonstrate that inhalation noises before stretches of speech are louder than silent breathing, and that they are louder before longer utterances than before shorter ones (< 1 s). We thus replicate the results from our earlier study which used close-talking head-mounted microphones, while avoiding the associated data loss due to cross-talk. We also show that inhalations are louder within than before a speaking turn. Hence, the study provides another piece of evidence in favour of communicative functions of respiratory noises serving as potential turn-taking cues.

1. Introduction

While audible breathing is commonly associated with chronic pulmonary disorders, heavy exertion or being a former Jedi knight gone bad, inhalation and exhalation noises occurring in speech are louder than we are normally aware of. Any recording of spontaneous non-pathological speech provides ample evidence for this seemingly pedestrian observation.

However, in spite of the perceptual salience of respiratory noises, relatively little is known about their pragmatics. They have been shown to improve recall of synthetic speech (Whalen, Hoequist & Sheffert, 1995), express dispreference (Kendrick & Torreira, 2015) and emotion (Yuan & Li, 2007), and they have also been shown to mark thematic structure of read texts (Bailly & Gouvernayre, 2012). In addition, we recently presented some evidence based on detection thresholds for pauses accompanied and...

You are not authenticated to view the full text of this chapter or article.

This site requires a subscription or purchase to access the full text of books or journals.

Do you have any questions? Contact us.

Or login to access all content.