Whispered Speech Reveals Articulatory Control for Voicing Contrasts

Category: Modelling · Effect: Strong effect · Year: 2009

Maintaining distinct glottal aperture differences for voiced versus voiceless consonants even in whispered speech suggests that motor control targets for voicing are primarily articulatory, not purely acoustic.

Design Takeaway

When designing systems that interact with or generate speech, consider that the underlying motor control for phonetic distinctions is likely rooted in physical articulation, not just acoustic output.

Why It Matters

Understanding the underlying motor control mechanisms in speech production is crucial for developing more sophisticated speech synthesis and recognition systems. This research provides evidence that designers of such systems should consider articulatory constraints when modelling phonetic contrasts.

Key Finding

Even when speaking without vocal fold vibration (whispering), the physical adjustments in the larynx to distinguish between voiced and voiceless sounds remain, indicating that the brain is controlling the physical movements of the vocal apparatus rather than just the resulting sound.

Key Findings

Research Evidence

Aim: To investigate whether speech motor control targets for voicing contrasts are primarily articulatory or acoustic by examining whispered speech production.

Method: Experimental research

Procedure: The study involved two experiments. The first used endoscopic video recordings of the larynx during the production of phonological voicing contrasts in normal and whispered speech. A perceptual experiment was conducted on the elicited speech material to gather listener judgments.

Context: Speech production and motor control research

Design Principle

Articulatory control is a fundamental aspect of speech production that influences acoustic outcomes.

How to Apply

When developing text-to-speech engines, ensure the underlying articulatory models accurately reflect the physical differences required for phonemic distinctions, even in reduced speech modes like whispering.

Limitations

The study focused on specific phonetic contrasts and may not generalize to all speech sounds or languages. The perceptual experiment details were not fully elaborated in the abstract.

Student Guide (IB Design Technology)

Simple Explanation: When we speak, our brains control the muscles in our mouths and throats. This study shows that even when we whisper, the way we move those muscles to make sounds like 'd' versus 't' is different, suggesting our brains are telling our bodies how to move, not just what sound to make.

Why This Matters: Understanding how humans control their speech can help you design better voice interfaces, speech recognition systems, or even assistive communication devices.

Critical Thinking: If articulatory targets are primary, what are the implications for designing systems that need to adapt to diverse vocal tract anatomies or speech impediments?

IA-Ready Paragraph: Research into speech motor control, such as Mills (2009), suggests that phonetic distinctions are maintained through articulatory targets. This implies that when designing systems involving speech, prioritizing the modelling of physical articulation over purely acoustic output can lead to more robust and naturalistic results.

Project Tips

How to Use in IA

Examiner Tips

Independent Variable: Speech condition (normal vs. whispered), Phonetic contrast (voiced vs. voiceless obstruent)

Dependent Variable: Glottal aperture, Perceptual judgments

Controlled Variables: Phonological voicing contrasts, Obstruents

Strengths

Critical Questions

Extended Essay Application

Source

Speech motor control variables in the production of voicing contrasts and emphatic accent · ERA · 2009