Title :
Towards ASR on partially corrupted speech
Author :
Hennansky, H. ; Tibrewala, Sangita ; Pave, Misha
Author_Institution :
Oregon Graduate Inst. of Sci. & Technol., Beaverton, OR, USA
Abstract :
A new highly parallel approach to automatic recognition of speech, inspired by early Fetcher´s research on articulation index, and based on independent probability estimates in several sub-bands of the available speech spectrum, is presented. The approach is especially suitable for situations when part of the spectrum of speech is computed. In such cases, it can yield an order-of-magnitude improvement in the error rate over a conventional full-band recognizer
Keywords :
probability; speech recognition; articulation index; automatic speech recognition; error rate; full-band recognizer; highly parallel approach; independent probability estimates; partially corrupted speech; speech spectrum; Automatic speech recognition; Decoding; Degradation; Frequency; Humans; Merging; Signal processing; Speech enhancement; Speech recognition; Viterbi algorithm;
Conference_Titel :
Spoken Language, 1996. ICSLP 96. Proceedings., Fourth International Conference on
Conference_Location :
Philadelphia, PA
Print_ISBN :
0-7803-3555-4
DOI :
10.1109/ICSLP.1996.607154
