JOURNAL ARTICLE

Speech Recognition Accuracy Prediction Using Speech Quality Measure

Seung-eun JiWooil Kim

Year: 2016 Journal:   The Journal of the Korean Institute of Information and Communication Engineering Vol: 20 (3)Pages: 471-476

Abstract

본 논문에서는 음성 특성 지표를 이용한 음성 인식 성능 예측 실험의 내용을 소개한다. 선행 실험에서 효과적인 음성 인식 성능 예측을 위해 대표적인 음성 인식 성능 지표인 단어 오인식률과 상관도가 높은 여러 가지 특성 지표들을 조합하여 새로운 성능 지표를 제안하였다. 제안한 지표는 각 음성 특성 지표를 단독으로 사용할 때 보다 단어 오인식률과 높은 상관도를 나타내 음성 인식 성능을 예측하는데 효과적임을 보였다. 본 실험에서는 이 결과를 근거하여 조합에 사용된 음성 특성 지표를 채택하여 4차원 특징 벡터를 생성하고 GMM 기반의 음성 인식 성능 예측기를 구축한다. 가우시안 요소를 증가시키며 실험한 결과 제안된 시스템은 babble 잡음, 자동차 잡음에서 모두 SNR이 낮을수록 단어 오인식률을 높은 확률로 예측함을 확인하였다. This paper presents our study on speech recognition performance prediction. Our initial study shows that a combination of speech quality measures effectively improves correlation with Word Error Rate (WER) compared to each speech measure alone. In this paper we demonstrate a new combination of various types of speech quality measures shows more significantly improves correlation with WER compared to the speech measure combination of our initial study. In our study, SNR, PESQ, acoustic model score, and MFCC distance are used as the speech quality measures. This paper also presents our speech database verification system for speech recognition employing the speech measures. We develop a WER prediction system using Gaussian mixture model and the speech quality measures as a feature vector. The experimental results show the proposed system is highly effective at predicting WER in a low SNR condition of speech babble and car noise environments.

Keywords:
Speech recognition Measure (data warehouse) Computer science Quality (philosophy) Artificial intelligence Natural language processing Data mining

Metrics

1
Cited By
0.24
FWCI (Field Weighted Citation Impact)
3
Refs
0.53
Citation Normalized Percentile
Is in top 1%
Is in top 10%

Citation History

Topics

Speech and Audio Processing
Physical Sciences →  Computer Science →  Signal Processing
Music and Audio Processing
Physical Sciences →  Computer Science →  Signal Processing
Speech Recognition and Synthesis
Physical Sciences →  Computer Science →  Artificial Intelligence

Related Documents

JOURNAL ARTICLE

Speech Recognition Accuracy Measure using Deep Neural Network for Effective Evaluation of Speech Recognition Performance

Seung-eun JiWooil Kim

Journal:   The Journal of the Korean Institute of Information and Communication Engineering Year: 2017 Vol: 21 (12)Pages: 2291-2297
JOURNAL ARTICLE

Enhancement Quality and Accuracy of Speech Recognition System Using Multimodal Audio-Visual Speech signal

Eslam ElmaghrabyAmr M. GodyMohamed Hesham Farouk

Journal:   The Egyptian Journal of Language Engineering /The Egyptian Journal of Language Engineering Year: 2017 Vol: 4 (2)Pages: 27-40
JOURNAL ARTICLE

Single-Ended Speech Quality Prediction Based on Automatic Speech Recognition

Rainer HüberJasper OosterBernd Meyer

Journal:   Journal of the Audio Engineering Society Year: 2018 Vol: 66 (10)Pages: 759-769
JOURNAL ARTICLE

Robust Speech Feature Prediction Using Mel-LPC to Improve Recognition Accuracy

S. LokeshG. Balakrishn

Journal:   Information Technology Journal Year: 2012 Vol: 11 (11)Pages: 1644-1649
© 2026 ScienceGate Book Chapters — All rights reserved.