| Autor: |
Vishrut Nayak, Ramesh M. Kagalkar, Kashinath, Suyodh Kittur, Shashank Simpi |
| Rok vydání: |
2020 |
| Předmět: |
|
| Zdroj: |
Learning and Analytics in Intelligent Systems ISBN: 9783030469429 |
| DOI: |
10.1007/978-3-030-46943-6_17 |
| Popis: |
Language is an important way of communication for people and is expressed in both written and speech form. The idea of this paper is on computing the framework for speech recognition wherever information required is to be made available in short time span. The input taken is in spoken audio form and the output derived is text description. In this system analysis of audio content is performed and then classified into text description. It has two sections, Training and Testing, Where in training section, the process of feature extraction of audio content is performed and are stored in database, in the testing section matches the features and provides the text description of the audio content. Further by applying topic modelling method and the summary of audio file is derived. This system uses database of 50 domain audio samples, with each domain contains 5 audio samples. The performance of the system is measured in terms of processing time and recognition. This system is considered to be high accuracy rate as compared to conventional method. Hence we derived accuracy rate of the system is up to 89.12%. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
