• Journal of KIISE
• /
• v.44 no.7
• /
• pp.674-679
• /
• 2017

In big-data environments wherein large amounts of text documents are produced daily, titles are very important clues that enable a prompt catching of the key ideas in documents; however, titles are absent for numerous document types such as blog articles and social-media messages. In this paper, a title-generation model for which sequence-to-sequence RNNs with attention and copying mechanisms are employed is proposed. For the proposed model, input sentences are encoded based on bi-directional GRU (gated recurrent unit) networks, and the title words are generated through a decoding of the encoded sentences with keywords that are automatically selected from the input sentences. Regarding the experiments with 93631 training-data documents and 500 test-data documents, the attention-mechanism performances are more effective (ROUGE-1: 0.1935, ROUGE-2: 0.0364, ROUGE-L: 0.1555) than those of the copying mechanism; in addition, the qualitative-evaluation radiative performance of the former is higher.

• Phonetics and Speech Sciences
• /
• v.12 no.2
• /
• pp.29-37
• /
• 2020

Recently, the neural network-based deep learning algorithm has dramatically improved performance compared to the classical Gaussian mixture model based hidden Markov model (GMM-HMM) automatic speech recognition (ASR) system. In addition, researches on end-to-end (E2E) speech recognition systems integrating language modeling and decoding processes have been actively conducted to better utilize the advantages of deep learning techniques. In general, E2E ASR systems consist of multiple layers of encoder-decoder structure with attention. Therefore, E2E ASR systems require data with a large amount of speech-text paired data in order to achieve good performance. Obtaining speech-text paired data requires a lot of human labor and time, and is a high barrier to building E2E ASR system. Therefore, there are previous studies that improve the performance of E2E ASR system using relatively small amount of speech-text paired data, but most studies have been conducted by using only speech-only data or text-only data. In this study, we proposed a semi-supervised training method that enables E2E ASR system to perform well in corpus in different domains by using both speech or text only data. The proposed method works effectively by adapting to different domains, showing good performance in the target domain and not degrading much in the source domain.

• Journal of Internet of Things and Convergence
• /
• v.9 no.3
• /
• pp.87-92
• /
• 2023

As times develop, information becomes more diverse and methods of obtaining it become more diverse. About 80% of the amount of information gained in life is acquired through the visual sense. However, visually impaired people have limited ability to interpret visual materials. That's why Braille, a text for the blind, appeared. However, the Braille decoding rate of the blind is only 5%, and as the demand of the blind who want various forms of platforms or materials increases over time, development and product production for the blind are taking place. An example of product production is braille books, which seem to have more disadvantages than advantages, and unlike non-disabled people, it is true that access to information is still very difficult. In this paper, we designed a CNN-based Braille conversion and voice output device to make it easier for visually impaired people to obtain information than conventional methods. The device aims to improve the quality of life by allowing books, text images, or handwritten images that are not made in Braille to be converted into Braille through camera recognition, and designing a function that can be converted into voice according to the needs of the blind.