The Rated L2 Speech Corpus
This is a corpus of speech recorded by 28 learners of English, from a variety of L1 backgrounds including Arabic, Chinese, French, Korean, Spanish, Tai, and Turkish. Subjects each saw a question, were given some seconds to read the question, then were asked to answer the question orally. Each waveform was transcribed, and its second-language fluency was rated. For more information: Su-Youn Yoon, Lisa Pierce, Amanda Huensch, Eric Juul, Samantha Perkins, Richard Sproat, and Mark Hasegawa-Johnson, Construction of a rated speech corpus of L2 learners' speech, CALICO journal, 2009.