Inseon Jang

Development of Multichannel Sound Scene Visualization Tool with MPEG Surround Multichannel Decoder

In this paper, a multichannel sound scene visualization tool is designed and implemented based on MPEG surround multichannel decoder. The aim of this tool is to effectively display frequency dependent dynamic characteristics of multichannel sound scene. It uses only spatial cue parameters extracted from MPEG Surround bitstream without additional computations such as T/F transform.

The MPEG Interactive Music Application Format Standard [Standards in a Nutshell]

The music industry is going through a transformation, and new interactive music services have emerged. It is envisaged that this new concept of digital music content will dominate the next generation of music services. A standardized file format is inevitably required to provide the interoperability between various interactive music players and interactive music albums. This issue is addressed in a new standard by the Moving Picture Experts Group (MPEG), known as the MPEG-A Interactive Music Application Format (IM AF). IM AF integrates multiple audio tracks with appropriate additional information, enabling users to experience various preset mixes and to make their own mixes complying with interactivity rules imposed by the music composers with the aim of fitting their artistic creation.

Why blind source separation for blind equalization of multiple channels

The paper addresses a blind source separation (BSS)-based method for blind equalization (BE) of single-input-multiple-output (SIMO) channels. We show here that a BSS-based BE method has some advantages over other methods. These include: (1) robustness in channel length mismatch; (2) reduction of noise effect (with a selected equalizer delay); (3) robustness for ill-conditioned channels. We illustrate that a minimum norm selection scheme achieves these advantages in the framework of BSS-based BE. Numerical experiments confirm the high performance of our proposed method.

Dynamic Subtitle Authoring Method Based on Audio Analysis for the Hearing Impaired

The broadcasting and the Internet are important parts of modern society that a life without media is now unimaginable. However, hearing impaired people have difficulty in understanding media content due to the loss of audio information. If subtitles are available, subtitling with video can be helpful. In this paper, we propose a dynamic subtitle authoring method based on audio analysis for the hearing impaired. We analyze the audio signal and explore a set of audio features that include STE, ZCR, Pitch and MFCC. Using these features, we align the subtitle with the speech and match extracted speech features to subtitle as different text colors, sizes and thicknesses. Furthermore, it highlights the text via aligning them with the voice and tagging the speaker ID using the speaker recognition.

Survey and Analysis of the Audio Description Acceptance for Improving the Media Accessibility of the Visually Impaired

For people with physical or sensory limitations, broadcasting is the main means of information acquisition and leisure. Recently, changes in the media environment, such as convergence of broadcasting and communication, digital·mobile conversion of broadcasting, and active media usage behavior of users, make broadcasting accessibility of the disabled difficult, and as a result, the information gap between the disabled and the non-disabled is increasing. A notice on broadcasting rights for the disabled was enacted in consequence of the amendment of the Broadcasting Law in July 2011 and the web accessibility guideline became more effective with the amendment of the National Informatization Act in 2013 so that legal basis for the right of media access for the disabled was established. However, media services for them are still lacking quantitatively and qualitatively. In this study, we describe the present status of the audio description service for the visually impaired, and analyze the results of the questionnaire survey on the usage status, satisfaction and improvement requirements of the audio description service for 100 visually impaired people. Keyword : Audio Description, Descriptive Video Service, Broadcasting for the disabled Copyright C 2017 Korean Institute of Broadcast and Media Engineers. All rights reserved. “This is an Open-Access article distributed under the terms of the Creative Commons BY-NC-ND (http://creativecommons.org/licenses/by-nc-nd/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited and not altered.” 일반논문 (Regular Paper) 방송공학회논문지 제22권 제2호, 2017년 3월 (JBE Vol. 22, No. 2, March 2017) https://doi.org/10.5909/JBE.2017.22.2.214 ISSN 2287-9137 (Online) ISSN 1226-7953 (Print) 장인선 외 4인: 시각장애인 미디어접근권 향상을 위한 해설오디오 수용도 조사 및 분석 215 (Inseon Jang et al.: Survey and Analysis of the Audio Description Acceptance for Improving the Media Accessibility of the Visually Impaired)

A pitch-synchronous speech analysis and synthesis method for DNN-SPSS system

This paper proposes a pitch-synchronous deep neural network (DNN)-based statistical parametric speech synthesis (SPSS) system. The pitch-synchronous frames defined by the locations of glottal closure instants (GCIs) are used to extract speech parameters, which significantly reduce coupling effects between vocal tract and excitation signals. As a result, the distribution of spectral parameters within the same context of phonetic classes becomes more uniform, which improves a model trainability especially for a small-scaled DNN framework. Although the effectiveness of pitch-synchronous approach has been proven in other applications, it is not trivial to integrate the method into the typical DNN-based SPSS systems that have regularized structures, i.e. fixed frame rate and fixed dimension of features. In this paper, we design a new DNN-based SPSS system that pitch-synchronously trains and generates speech parameters. Objective and subjective test results verify the superiority of the proposed system compared to the conventional approach.

Semi-automatic DVS Authoring Method

Descriptive video service (DVS) is the main method of making programs accessible to those with seeing disabilities, but only a few of conventional broadcasting programs have been reproduced in the form of DVS contents because of practical limitations. It takes much of the time and professional manpower to produce the DVS contents so it is quite costly. In this paper, we propose semi-automatic DVS authoring method. Non-dialog sections detected through audio/subtitles analysis are recommended and then the author is able to insert appropriate audio description (AD) scripts and to produce their synthesized AD using TTS easily. Currently we have completed a basic study and developed the trial version of the proposed.

Interactive music contents authoring system based on IM AF

This paper describes an authoring system, which can easily and conveniently produce an interactive music contents by using MPEG-A IM AF (Interactive Music Application Format) standard. Interactive music service provides a user interactivity to music contents, which means that the user can recompose the music contents as his/her preference. Detailed implementation of the interactive music authoring system based on IM AF and its capabilities are presented along with usage examples.

F-SEONS: a second-order frequency-domain algorithm for noisy convolutive source separation

We present a frequency-domain method of noisy convolutive source separation, where we extend the SEONS algorithm (Choi et al. (2002)) that jointly exploits the nonstationarity and temporal structure of sources. Thus, the method is called F-SEONS, implying frequency-domain SEONS. Unlike most of the existing methods of convolutive source separation, we consider the case of noisy data and show that our F-SEONS algorithm identifies multivariate FIR channels in a robust way. In addition, we employ an H/sub /spl infin// filtering method in order to further suppress the sensor noise. Numerical experiments compared to other methods confirm the high performance of our proposed method.

F-SEONS: a second-order frequency domain algorithm for convolutive source separation in noisy environments [speech recognition applications]

Convolutive source separation is a problem of restoring original unknown sources from multiple sensor signals that are modelled as convolutive mixtures of mutually independent sources. It is a fundamental problem which plays a critical role in cocktail party speech recognition. Most of existing methods neglected the effect of sensor noise in the model of convolved mixtures. In this paper, we consider the case of noisy data and present a method of source separation, F-SEONS, which jointly exploits the nonstationarity and temporal structure of sources in the frequency-domain. Numerical experiments, comparing to other methods, confirm the high performance of our proposed method.