Work

A semi-automatic pipeline for transcribing and segmenting child speech

Year: 2025

Type: conference-paper

Abstract: This study evaluates both automated transcription (WhisperX) and forced alignment (MFA) in developing a semi-automated pipeline for obtaining acoustic vowel measures from field recordings from 275 children speaking a non-standard, English dialect, Scottish English. As expected, manual correction of speech transcriptions before forced alignment improves the quality of acoustic vowel measures with respect to manually-annotated data, though speech style and recording environment present some challenges for both tools. Adaptation of the MFA pre-trained english_us_arpa acoustic model towards the children's speech also improves the quality of acoustic measures, though greater improvement was not found by increasing training sample size.

Source: Interspeech (USB)

DOI: 10.21437/interspeech.2025-1030

Authors: Polychronia Christodoulidou, James Tanner, Jane Stuart-Smith, Michael McAuliffe, Mridhula Murali, Amy Smith, Lauren Taylor, Joanne Cleland, Anja Kuschmann

Language: English

Topic: Speech Recognition and Synthesis

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: Economic and Social Research Council

Award: Variability in Child Speech (VariCS)

OpenAlex ID: W4415433452

FWCI:

Cites:

Cited by:

Related to: