Transcribing disordered speech: the segmental and prosodic layers

Area of Science:

• Phonetics and Speech Science
• Linguistics
• Speech-Language Pathology

Background:

• Transcription is crucial for analyzing speech, especially disordered speech.
• Existing transcription methods face challenges with atypical speech production and prosodic features.
• Transcriber reliability and the extension of symbol sets are key practical issues.

Purpose of the Study:

• To explore the speech layers within a transcription toolkit.
• To discuss the theory and practice of phonetic transcription, focusing on narrow transcription for disordered speech.
• To investigate instrumental approaches and extend symbol sets for atypical speech.

Main Methods:

• Discussion of transcription theory, including broad vs. narrow phonetic transcription.
• Examination of transcriber reliability measurement.
• Exploration of extended symbol sets and instrumental approaches for atypical speech.
• Analysis of both segmental and prosodic transcription with sample illustrations.

Main Results:

• Narrow phonetic transcription is vital for disordered speech.
• Instrumental approaches and extended symbol sets can aid in transcribing atypical speech.
• Less consensus exists on prosodic transcription compared to segmental transcription, particularly for disorders.

Conclusions:

• The study provides a comprehensive overview of speech transcription layers.
• It underscores the importance of narrow transcription and instrumental methods for disordered speech.
• Further research is needed to standardize prosodic transcription and notation for disorders.