| Autor: |
M. Banu priya, E. Kannamal |
| Rok vydání: |
2020 |
| Předmět: |
|
| Zdroj: |
2020 International Conference on Computer Communication and Informatics (ICCCI). |
| DOI: |
10.1109/iccci48352.2020.9104068 |
| Popis: |
In recent years the communication between the humans got reduced. Human to machine communication got evolved and it is used in many application. Speech recognition system doesn’t use the words we say directly to perform the task. First it converts the words which we have spoken into text. Thus speech recognition system is also called as "automatic speech recognition" or "speech to text" translator. In this paper, HMM, DTD and ANS algorithms have been discussed. The simple model to convert speech to text along with deep learning and Natural Language Processing(NLP) is also implemented in this article. Hidden Markov Model(HMM) is also called probabilistic graphical model. In this model the sequence of hidden variables are identified by using the observed variables. Dynamic Time distorting (DTD), used to find the similarity between the two temporal sequences. The speed and time of the sequence may vary. This algorithm is mainly developed for speech recognition. ANS is originated on the neural structure of brain. The simple processing units are connected through a complex communication network. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
