Effects of syllable-final duration variation on the identification of synthetic /ba/-/wa/ and /bas/-/was/ speech continua by humans and birds

• Published in: The Journal of the Acoustical Society of America 2007-11, Vol.122 (5_Supplement), p.2973-2973
• Main Authors: Welch, Thomas E., Dent, Micheal L., Sawusch, James R.
• Format: Article
• Language: English
• ISSN: 0001-4966; 1520-8524
• DOI: 10.1121/1.2942606

In an attempt to test whether experience with or knowledge of language is necessary to show typical human only duration-based tendencies in speech perception tasks, budgerigars (Melopsittacus undulatus) and zebra finches (Taeniopygia guttata) were trained to categorize the endpoint stimuli of four synthetic continua consisting of CV and CVC stimuli, with both short and long syllable-final durations. Humans and birds were then tested on the exact same sets of full ten-step continua and their performance was compared in terms of their identification responses and of phoneme boundaries along the /b/-/w/ stop-glide continua. Operant conditioning was used to train the birds to respond differentially to /b/ and /w/ stimuli. Overall, zebra finches, who unlike budgerigars, cannot mimic human speech, were poor at categorizing the human speech tokens and very few could even categorize two endpoints above criterion levels to be moved onto testing. The birds that were able to learn the task showed similar identification results to humans for the continua. The implications of these results and comparisons of the effects of phoneme duration (as in human speaking rate) between humans and birds will be discussed. For Speech Communication Best Student Paper Award Best ‘‘Outstanding Paper By A Young Presenter’’ Awards.