Publication Details
TRAP-based Techniques for Recognition of Noisy Speech
TRAP techniques, noisy speech recognition, multistream processing, feature combination
Paper persents performance of TRAP tecniques on noisy speech data, showing improvement using separate information processing with following combination of processed streams.
This paper presents a systematic study of performance of TempoRAl Patterns (TRAP) based features and their proposed modifications and combinations for speech recognition in noisy environment. The experimental results are obtained on AURORA2 database with clean training data. We observed large dependency of performance of different TRAP modifications on noise level. Earlier proposed TRAP system modifications help in clean conditions but degrade the system performance in presence of noise. The combination techniques on the other hand can bring large improvement in case of weak noise and degrade only slightly for strong noise cases. The vector concatenation combination technique is improving the system performance up to strong noise.
@INPROCEEDINGS{FITPUB8434, author = "Franti\v{s}ek Gr\'{e}zl and Jan \v{C}ernock\'{y}", title = "TRAP-based Techniques for Recognition of Noisy Speech", pages = "270--277", booktitle = "Proc. 10th International Conference on Text Speech and Dialogue (TSD 2007)", series = "LNCS", year = 2007, location = "Berlin, DE", publisher = "Springer Verlag", ISBN = "978-3-540-74627-0", language = "english", url = "https://www.fit.vut.cz/research/publication/8434" }