Title :
Statistical speech reconstruction at the phoneme level
Author :
Savic, Michael ; Moore, Michael D. ; Scoville, Christopher
Author_Institution :
ECSE Dept., Rensselear Polytech. Inst., New York, NY, USA
Abstract :
Statistical methods for reconstructing speech at the phoneme level are used to find missing phonemes that are removed from sentences in the TIMIT corpus. Probabilities for the occurrence of the missing phoneme(s) are generated and the most likely candidate(s) selected to reconstruct the sentence. The method includes symmetric and asymmetric ´confidence windowing´ around the missing phoneme(s) for determination of the most likely candidates. The reconstruction rates for one or more phonemes missing in a sequence can exceed 85%
Keywords :
probability; signal reconstruction; speech processing; statistical analysis; TIMIT corpus; asymmetric confidence windowing; missing phonemes probability; phoneme level speech reconstruction; phoneme reconstruction rates; sentence reconstruction; statistical speech reconstruction; symmetric confidence windowing; Databases; Economic indicators; Interference; Natural languages; Oral communication; Probability; Signal processing; Speech; Statistical analysis;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2001. Proceedings. (ICASSP '01). 2001 IEEE International Conference on
Conference_Location :
Salt Lake City, UT
Print_ISBN :
0-7803-7041-4
DOI :
10.1109/ICASSP.2001.940917
