Speech Perception Survey
|Submitter Email:||click here to access email|
Thank you to all of you who have replied to my query!
I have received quite a few messages from folks offering help. Many people, including Isabelle Buchstaller, suggested LDC as a great source of spoken word corpora. However, LDC subscriptions are costly. I have looked at the descriptions of some of the LDC corpora
and, unfortunately, many of them contained telephone channel speech. While this is great for discourse analysis, the bandwidth and dynamic range of the telephone channel make such recordings very unreliable for acoustic analysis.
Joe Picone was extremely kind and sent me a few samples of digitally recorded 16-bit, 16,000 Hz word list samples of Southern Speech. This was exactly what I needed. I would like to extend my gratitude to Prof. Picone for sending me those files.
Malcah Yaeger-Dror made a few very interesting remarks concerning the differences between word list and conversational data:
''First off, you might get some southern speech from Dennis Preston, who's right there. I have some 'snippets' that i downloaded from the radio that have monophthongization of (ay) more than what you have in those word lists [although the last two guys do monophthongize variably].''
Of course, I agree that the monophthongization of /ay/ may occur more often in casual speech. I will look into such data and try to compare those two different sets of data.
Alicia Spiegel, who studies pidgin/Creole languages, suggested that Walt Wolfram might have interesting data to share. I will look into that. Walt Wolfram has headed quite a few research projects in the American South, as well as on the islands off the coast of NC. I will write to him and ask about his collections.
My goal was to obtain several high quality Southern samples and to find out about repositories of Southern Speech. Thanks to your help, I was able to get exactly what I needed. Thank you!
|Original Query:||Read original query|
Sums main page