Loading...
Please wait, while we are loading the content...
Similar Documents
A comparison of DFT, PLP and Cochleagram for alphabet recognition (1991)
| Content Provider | CiteSeerX |
|---|---|
| Author | Fanty, Mark Cole, Roiiald |
| Description | The English alphabet is a small but dificult vocab-ulary fo r speech recognition, with many fine phonetic distinctions, such as M / N and B/V. We use speaker-independent classification of isolated English letters t o evaluate the relative performance of the D F T, Percep-tual Linear Predictive analysis, and the cochleagram auditory model. Feedforward neural network classi-fiers were trained using all three representations o n 60 speakers and tested on 60 new speakers. Training and testing data were independently modified b y adding two levels of Gaussian noise and babble (20 random let-ter utterances, attenuated and given random offsets). PLP gave the best results, especially when trained or tested on Gaussian noise. 1 |
| File Format | |
| Language | English |
| Publisher Date | 1991-01-01 |
| Publisher Institution | In Proceedings of the 25th Asilomar Conference on Signals, Systems and Computers |
| Access Restriction | Open |
| Subject Keyword | Relative Performance New Speaker Speaker-independent Classification Alphabet Recognition Many Fine Phonetic Distinction English Alphabet Gaussian Noise Feedforward Neural Network Classi-fiers Isolated English Letter Dificult Vocab-ulary Fo Speech Recognition Cochleagram Auditory Model Random Offset Percep-tual Linear Predictive Analysis Let-ter Utterance |
| Content Type | Text |
| Resource Type | Article |
