Utterances in infant-directed speech are shorter, not slower.

Area of Science:

• Linguistics
• Developmental Psychology
• Speech Science

Background:

• Infant-directed speech (IDS) is widely believed to be slower than adult-directed speech (ADS).
• Previous research often assumes a general slowing of articulation in IDS.
• The specific acoustic and prosodic mechanisms driving speech rate differences remain debated.

Purpose of the Study:

• To investigate the acoustic basis for the observed speech rate differences between IDS and ADS.
• To determine whether the slowing in IDS is uniform or driven by specific prosodic features.
• To examine the role of utterance length and phrase-final lengthening in IDS speech rate.

Main Methods:

• Analysis of acoustic recordings from 22 Japanese mothers speaking to their infants and to adults.
• Measurement of speech rate, including mean speech rate and durational properties of speech segments.
• Focus on phrase-final lengthening and its correlation with utterance length and speech rate.

Main Results:

• IDS exhibited a lower mean speech rate compared to ADS.
• This difference was not due to uniform slowing of all speech sounds.
• The reduced speech rate in IDS was primarily driven by increased phrase-final lengthening.
• Phrase-final lengthening effects were disproportionately larger in IDS due to shorter utterance durations.

Conclusions:

• The speech rate difference between IDS and ADS is primarily an artifact of prosodic structure, specifically phrase-final lengthening.
• Shorter utterances in IDS amplify the impact of phrase-final lengthening, leading to a slower overall rate.
• Accurate studies of speech rate in IDS must consider utterance-internal prosodic characteristics.