Robust Speech Recognition in the presence of noise using medical data

被引：0

作者：

Athanaselis, Theologos ^{[1
]}

Bakamidis, Stelios ^{[1
]}

Giannopoulos, George ^{[1
]}

Dologlou, Ioannis ^{[1
]}

Fotinea, Evita ^{[1
]}

机构：

[1] Inst Language & Speech Proc, Dept Speech Technol, Athens, Greece

来源：

2008 IEEE INTERNATIONAL WORKSHOP ON IMAGING SYSTEMS AND TECHNIQUES | 2008年

关键词：

voice control MRI; Speech Recognition system; Non-linear Spectral Subtraction; Truncated SVD Procedure;

D O I：

暂无

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

This paper discusses the improvement of speech recognition in the presence of noise, when a parametric method of signal enhancement is used. The speech enhancement method improves the performance of voice control MRI. This is important since errors in the presence of noise are more frequent and tend to make applications such as spoken dialogue systems, too cumbersome to use. The input signal is corrupted with MRI noise with varying signal-to-noise ratio. A non-linear spectral subtraction method (NSS) as well as an SVD based noise reduction techniques (ISE) are used in conjunction with the Speech Recognition system of the FAST project, to quantify the impact of speech enhancement.

引用

页码：347 / 350

页数：4

共 11 条

[1] ATHANASELIS T, SIGNAL ENHANCEMENT C
[2] Boersma P., 1993, Accurate short-term analysis of the fundamental frequency and the harmonics-to-noise ratio of a sampled sound, P97, DOI DOI 10.1371/JOURNAL.PONE.0069107
[3] SPEECH ENHANCEMENT FROM NOISE - A REGENERATIVE APPROACH
DENDRINOS, M
BAKAMIDIS, S
CARAYANNIS, G
[J]. SPEECH COMMUNICATION, 1991, 10 (01) : 45 - 57
[4] DOCLO S, 1998, P 5 INT C SPOK LANG, P1435
[5] FOTINEA SE, EMOTION SPEECH INTEG
[6] Kamppari SO, 2000, INT CONF ACOUST SPEE, P1799, DOI 10.1109/ICASSP.2000.862103
[7] KYRIAKOU C, 2001, P 4 EUR C NOIS CONTR, V2, P702
[8] LIPMMAN RP, 1997, SPEECH COMMUN, V22, P1
[9] PELLOM BL, 1997, P IEEE MIDW S CIRC S, P125
[10] UHL C, 2001, P IEEE ICASSP SALT L, V1, P281

← 1 2 →