A PC-based neural network for recognition of difficult syllables using LPC coefficient difference /
In this article we investigated the recognition of difficult CV(consonant-vowel) and VC(vowel-consonant) syllables using linear prediction coefficients and a PC-based neural network approach. The speech corpus consisted of41 syllables produced by three speakers in three different vowel contexts.The...
| Main Authors: | , , |
|---|---|
| Format: | |
| Subjects: |
| _version_ | 1826433519209938944 |
|---|---|
| author | 348610 Shim, C. Espinoza-Varas, B. Cheung, J. Y. |
| author_facet | 348610 Shim, C. Espinoza-Varas, B. Cheung, J. Y. |
| author_sort | 348610 Shim, C. |
| collection | OCEAN |
| description | In this article we investigated the recognition of difficult CV(consonant-vowel) and VC(vowel-consonant) syllables using linear prediction coefficients and a PC-based neural network approach. The speech corpus consisted of41 syllables produced by three speakers in three different vowel contexts.The input to the neural network was the differences in linear predictioncoefficients sampled at each syllable's time-waveform. A fully connectedthree-layered back-propagation neural network was trained by the Deltalearning rule. With a relatively small number of parameters for eachsyllable for training, prelimanary results based on 123 tokens of 41difficult syllables spoken within a sentence context by three speakers(one male and two female) indicated tha the recognition accuracy was ashigh as 96.8%. |
| first_indexed | 2024-03-05T07:47:04Z |
| format | |
| id | KOHA-OAI-TEST:385739 |
| institution | Universiti Teknologi Malaysia - OCEAN |
| last_indexed | 2024-03-05T07:47:04Z |
| record_format | dspace |
| spelling | KOHA-OAI-TEST:3857392020-12-19T17:13:27ZA PC-based neural network for recognition of difficult syllables using LPC coefficient difference / 348610 Shim, C. Espinoza-Varas, B. Cheung, J. Y. In this article we investigated the recognition of difficult CV(consonant-vowel) and VC(vowel-consonant) syllables using linear prediction coefficients and a PC-based neural network approach. The speech corpus consisted of41 syllables produced by three speakers in three different vowel contexts.The input to the neural network was the differences in linear predictioncoefficients sampled at each syllable's time-waveform. A fully connectedthree-layered back-propagation neural network was trained by the Deltalearning rule. With a relatively small number of parameters for eachsyllable for training, prelimanary results based on 123 tokens of 41difficult syllables spoken within a sentence context by three speakers(one male and two female) indicated tha the recognition accuracy was ashigh as 96.8%.In this article we investigated the recognition of difficult CV(consonant-vowel) and VC(vowel-consonant) syllables using linear prediction coefficients and a PC-based neural network approach. The speech corpus consisted of41 syllables produced by three speakers in three different vowel contexts.The input to the neural network was the differences in linear predictioncoefficients sampled at each syllable's time-waveform. A fully connectedthree-layered back-propagation neural network was trained by the Deltalearning rule. With a relatively small number of parameters for eachsyllable for training, prelimanary results based on 123 tokens of 41difficult syllables spoken within a sentence context by three speakers(one male and two female) indicated tha the recognition accuracy was ashigh as 96.8%.56PSZJBLNeural circuitrySpeech perception |
| spellingShingle | Neural circuitry Speech perception 348610 Shim, C. Espinoza-Varas, B. Cheung, J. Y. A PC-based neural network for recognition of difficult syllables using LPC coefficient difference / |
| title | A PC-based neural network for recognition of difficult syllables using LPC coefficient difference / |
| title_full | A PC-based neural network for recognition of difficult syllables using LPC coefficient difference / |
| title_fullStr | A PC-based neural network for recognition of difficult syllables using LPC coefficient difference / |
| title_full_unstemmed | A PC-based neural network for recognition of difficult syllables using LPC coefficient difference / |
| title_short | A PC-based neural network for recognition of difficult syllables using LPC coefficient difference / |
| title_sort | pc based neural network for recognition of difficult syllables using lpc coefficient difference |
| topic | Neural circuitry Speech perception |
| work_keys_str_mv | AT 348610shimc apcbasedneuralnetworkforrecognitionofdifficultsyllablesusinglpccoefficientdifference AT espinozavarasb apcbasedneuralnetworkforrecognitionofdifficultsyllablesusinglpccoefficientdifference AT cheungjy apcbasedneuralnetworkforrecognitionofdifficultsyllablesusinglpccoefficientdifference AT 348610shimc pcbasedneuralnetworkforrecognitionofdifficultsyllablesusinglpccoefficientdifference AT espinozavarasb pcbasedneuralnetworkforrecognitionofdifficultsyllablesusinglpccoefficientdifference AT cheungjy pcbasedneuralnetworkforrecognitionofdifficultsyllablesusinglpccoefficientdifference |