Talker-specific learning in speech perception

Area of Science:

• Auditory perception
• Speech processing
• Psycholinguistics

Background:

• Talker identity recognition is crucial for speech comprehension.
• Perceptual learning can enhance auditory processing skills.

Purpose of the Study:

• Investigate how learning talker identity affects spoken word and sentence recognition.
• Determine the influence of training stimuli (words vs. sentences) on voice learning generalization.
• Assess the impact of talker-specific information on linguistic content processing.

Main Methods:

• Three experiments involving listeners training to identify 10 talkers' voices.
• Listeners were trained using isolated words or sentence-length utterances.
• Intelligibility tests assessed recognition of spoken words and sentences produced by familiar and unfamiliar talkers in noise.

Main Results:

• Familiar talker voices improved word identification when trained with words.
• Voice learning from sentences did not generalize well to isolated word recognition.
• Voice learning from sentences improved sentence intelligibility for familiar talkers.

Conclusions:

• Perceptual learning of talker identity enhances speech intelligibility.
• Generalization of voice learning depends on the talker-specific information available during training and testing.
• Sensitivity to talker-specific cues impacts the perception of linguistic speech properties.