From Natural to Artificial Intelligence - Algorithms and Applications 2018
DOI: 10.5772/intechopen.80026
|View full text |Cite
|
Sign up to set email alerts
|

Convolutional Neural Networks for Raw Speech Recognition

Abstract: State-of-the-art automatic speech recognition (ASR) systems map the speech signal into its corresponding text. Traditional ASR systems are based on Gaussian mixture model. The emergence of deep learning drastically improved the recognition rate of ASR systems. Such systems are replacing traditional ASR systems. These systems can also be trained in end-to-end manner. End-to-end ASR systems are gaining much popularity due to simplified model-building process and abilities to directly map speech into the text wit… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
7
0
2

Year Published

2019
2019
2022
2022

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 30 publications
(11 citation statements)
references
References 36 publications
0
7
0
2
Order By: Relevance
“…The key point of this system is to develop a model for the categorisation of the peripheral blood cells utilising CNNs [33]. The peripheral blood cells are classified into eight classes such as neutrophils, basophils, eosinophils, monocytes, immature granulocytes, lymphocytes, platelets, and erythroblasts.…”
Section: Proposed Approachmentioning
confidence: 99%
“…The key point of this system is to develop a model for the categorisation of the peripheral blood cells utilising CNNs [33]. The peripheral blood cells are classified into eight classes such as neutrophils, basophils, eosinophils, monocytes, immature granulocytes, lymphocytes, platelets, and erythroblasts.…”
Section: Proposed Approachmentioning
confidence: 99%
“…The NNs are widely used by many researchers in many different applications such as robotics [2][3][4][5][6][7], speech recognition [8,9], human face recognition [10,11], medical applications [12][13][14], manufacturing [15,16], and economics [17,18].…”
Section: Introductionmentioning
confidence: 99%
“…Speech recognition was also proposed based on NN. In [8], convolutional NN-based acoustic model for raw speech signal was discussed. In this case, the relation between raw speech signal and phones in a data-driven manner was established.…”
Section: Introductionmentioning
confidence: 99%
“…Atualmente, sistemas de reconhecimento automático de fala (automatic speech recognition -ASR) operam com atributos de entrada provenientes da transformada de Fourier de curto termo (short-time Fourier transform -STFT) [1] ou com atributos provenientes diretamente do sinal de fala bruto no domínio do tempo [2]. Independente do modo de operação, a extração de atributosé uma etapa fundamental em qualquer sistema de ASR [2], [3].…”
Section: Introductionunclassified
“…Atualmente, sistemas de reconhecimento automático de fala (automatic speech recognition -ASR) operam com atributos de entrada provenientes da transformada de Fourier de curto termo (short-time Fourier transform -STFT) [1] ou com atributos provenientes diretamente do sinal de fala bruto no domínio do tempo [2]. Independente do modo de operação, a extração de atributosé uma etapa fundamental em qualquer sistema de ASR [2], [3]. Nesse contexto, ambos os modos de operação buscam identificar e reter apenas atributos que mais contribuem para a geração de um modelo acústico (MA) quê Enio dos Santos Silva e Rui Seara, LINSE-Laboratório de Circuitos e Processamento de Sinais, Departamento de Engenharia Elétrica e Eletrônica, Universidade Federal de Santa Catarina (UFSC), Florianópolis, SC, Brasil, emails: enio@linse.ufsc.br; seara@linse.ufsc.br.…”
Section: Introductionunclassified