Does Musical Experience Facilitate Phonetic Accommodation During Human-Robot Interaction?

Area of Science:

• Speech Science
• Second Language Acquisition
• Human-Robot Interaction

Background:

• Phonetic accommodation is a key aspect of communication, involving adjustments in speech production.
• Social robots are increasingly used in human-computer interaction, necessitating research into effective communication strategies.
• The role of musical training in second language (L2) phonetic acquisition remains an area of active investigation.

Purpose of the Study:

• To investigate the impact of musical training on phonetic accommodation in L2 English speakers interacting with a social robot.
• To explore the motivations and strategies underlying phonetic accommodation in this context.
• To compare phonetic accommodation patterns between musicians and non-musicians.

Main Methods:

• Recruited 15 L2 English speakers with musical training (musician group) and 15 without (nonmusician group).
• Conducted four conversational tasks with the social robot Furhat.
• Analyzed spectral and prosodic cues of participants' speech production before and after interactions to quantify phonetic accommodations.

Main Results:

• Both groups exhibited similar convergence in vowel formants toward the robot's speech.
• The musician group showed greater divergence in prosodic cues (mean fundamental frequency, duration) compared to the nonmusician group.
• No significant group differences were observed in vowel formant convergence.

Conclusions:

• Findings suggest individuals adapt their speech to enhance communication efficiency with robots, potentially using a special speech register.
• Musical training may enhance phonetic accommodation, particularly in prosodic aspects, possibly due to improved auditory attention and working memory.
• Results have implications for speech communication training and social robot design for therapeutic applications.