Speaker-independent isolated digit recognition: multilayer perceptrons vs. dynamic time warping /
Former experiments have shown the benefit of using specific multi-layer architectures,the so-called time dealy neural networks,for phoneme recognition(Waibel,Hanazawa,Hinton,Shikano, & Lang 1988). Similar experiments on a speaker-independent task were also performed on a small set of minimal pai...
| Main Author: | |
|---|---|
| Format: | |
| Subjects: |
| _version_ | 1796652862513610752 |
|---|---|
| author | Botton, L. |
| author_facet | Botton, L. |
| author_sort | Botton, L. |
| collection | OCEAN |
| description | Former experiments have shown the benefit of using specific multi-layer architectures,the so-called time dealy neural networks,for phoneme recognition(Waibel,Hanazawa,Hinton,Shikano, & Lang 1988). Similar experiments on a speaker-independent task were also performed on a small set of minimal pairs(bottou,1988). In this paper we focus on a speaker-independent,global word recognition task with time delay networks. We first describe these networks as away for learning feature extractors by constrained back-propagation.Such a time-delay network is shown to be capable of dealing with a near real-sizedproblem: French digit recognition.The results are discussed and compared,onthe same data sets,with those obtained with a classical time warping system. |
| first_indexed | 2024-03-04T14:26:54Z |
| format | |
| id | KOHA-OAI-TEST:39516 |
| institution | Universiti Teknologi Malaysia - OCEAN |
| last_indexed | 2024-03-04T14:26:54Z |
| record_format | dspace |
| spelling | KOHA-OAI-TEST:395162020-12-19T16:57:29ZSpeaker-independent isolated digit recognition: multilayer perceptrons vs. dynamic time warping / Botton, L. Former experiments have shown the benefit of using specific multi-layer architectures,the so-called time dealy neural networks,for phoneme recognition(Waibel,Hanazawa,Hinton,Shikano, & Lang 1988). Similar experiments on a speaker-independent task were also performed on a small set of minimal pairs(bottou,1988). In this paper we focus on a speaker-independent,global word recognition task with time delay networks. We first describe these networks as away for learning feature extractors by constrained back-propagation.Such a time-delay network is shown to be capable of dealing with a near real-sizedproblem: French digit recognition.The results are discussed and compared,onthe same data sets,with those obtained with a classical time warping system.Former experiments have shown the benefit of using specific multi-layer architectures,the so-called time dealy neural networks,for phoneme recognition(Waibel,Hanazawa,Hinton,Shikano, & Lang 1988). Similar experiments on a speaker-independent task were also performed on a small set of minimal pairs(bottou,1988). In this paper we focus on a speaker-independent,global word recognition task with time delay networks. We first describe these networks as away for learning feature extractors by constrained back-propagation.Such a time-delay network is shown to be capable of dealing with a near real-sizedproblem: French digit recognition.The results are discussed and compared,onthe same data sets,with those obtained with a classical time warping system.12PSZJBLSpeech perception |
| spellingShingle | Speech perception Botton, L. Speaker-independent isolated digit recognition: multilayer perceptrons vs. dynamic time warping / |
| title | Speaker-independent isolated digit recognition: multilayer perceptrons vs. dynamic time warping / |
| title_full | Speaker-independent isolated digit recognition: multilayer perceptrons vs. dynamic time warping / |
| title_fullStr | Speaker-independent isolated digit recognition: multilayer perceptrons vs. dynamic time warping / |
| title_full_unstemmed | Speaker-independent isolated digit recognition: multilayer perceptrons vs. dynamic time warping / |
| title_short | Speaker-independent isolated digit recognition: multilayer perceptrons vs. dynamic time warping / |
| title_sort | speaker independent isolated digit recognition multilayer perceptrons vs dynamic time warping |
| topic | Speech perception |
| work_keys_str_mv | AT bottonl speakerindependentisolateddigitrecognitionmultilayerperceptronsvsdynamictimewarping |