Automatic vocal effort detection for reliable speech recognition

Zelinka, Petr; Sigmund, Milan

doi:10.1109/mlsp.2010.5589174

Cited by 9 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In future work, we will test the algorithms under acoustic conditions that may affect their reliability, such as presence of non-stationary noise [24] and vocal effort fluctuations [25]. Our goal is to find an optimal of effective algorithms for the correct classification of voiced and unvoiced sounds in continuous Czech speech.…”

Section: Discussionmentioning

confidence: 99%

Comparison of methods for determining speech voicing based on tests performed on paired consonants and continuous speech

Malucha

Sigmund

2022

Journal of Electrical Engineering

Self Cite

View full text Add to dashboard Cite

Voicing is an important phonetic characteristic of speech. Each phoneme belongs to a group of either voiced or unvoiced sounds. We investigated and compared the performance of five algorithms widely used to estimate speech voicing. All algorithms were implemented in Matlab and tested on both short consonants and continuous speech. Phonetically paired consonants (voiced vs unvoiced) and parts of read speech from audio books were used in the experiments. The tuned harmonics-to-noise ratio method gave the best results in both situations, ie for consonants and continuous speech. Using this method, the overall voicing of Czech, Polish, Hungarian and English was investigated. Hungarian speech showed the highest proportion of voiced parts, approx. 75 %. In other languages, the proportion of voiced parts was around 70 %.

show abstract

Section: Discussionmentioning

confidence: 99%

Comparison of methods for determining speech voicing based on tests performed on paired consonants and continuous speech

Malucha

Sigmund

2022

Journal of Electrical Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…In future work, the long-term speech spectrum and extracted spectral parameters will be tested for their robustness to various factors affecting speech, such as emotions [7], physical fatigue [12], vocal effort [9,28], and others [4]. Furthermore, it will be useful to also investigate the influence of adverse acoustic conditions, especially non-stationary noises [29] due to their indirect influence on speech production.…”

Section: Discussionmentioning

confidence: 99%

Speaker Discrimination Using Long-Term Spectrum of Speech

Sigmund

2019

ITC

Self Cite

View full text Add to dashboard Cite

In this article, we investigate a specific long-term speech spectrum with respect to its use for speaker recognition. The long-term effect was satisfied by averaging short-term autocorrelation coefficients over the whole utterance. The long-term spectrum was calculated by means of second-order linear prediction using the average autocorrelation coefficients. First, speaker discriminability of 32 individual parameters was evaluated by combining spectral energy and spectral slope in eight different frequency bands covering the range 0−4 kHz (seven narrow nonoverlapping subbands and one band spanning over the full range). Then, four subbands with the most discriminative capability were selected for speaker recognition. These subbands involve the frequencies of 0−1.2 kHz in total. In the main experiments, text-independent speaker recognition based on relative Euclidean distance was performed in each single subband as well as in multiple 2 to 4 subbands applying two types of speech data, complete continuous speech and voiced part of the same speech. The voiced speech seems to be generally more effective for speaker recognition using the long-term speech spectrum. The best recognition rates, i.e. 91.7% on complete speech and 100% on voiced speech, were achieved in optimal paired subbands. The long-term speech spectrum can complement the traditional voice features.

show abstract

“…Frequency-domain effects caused by changing the vocal effort have implications for data-driven speech technology applications relying on short-time spectral features such as mel frequency cepstral coefficients (MFCCs). In particular, the performance of, e.g., automatic speech recognition (ASR) and speaker recognition systems will be affected by vocal effort mismatch between the training and recognition phase [5] [6] [7]. In order to avoid the performance degradation caused by this mismatch, a detection system is needed to aid the recognizer in choosing acoustic models that are most appropriate for the changed conditions [7].…”

Section: Introductionmentioning

confidence: 99%

“…In particular, the performance of, e.g., automatic speech recognition (ASR) and speaker recognition systems will be affected by vocal effort mismatch between the training and recognition phase [5] [6] [7]. In order to avoid the performance degradation caused by this mismatch, a detection system is needed to aid the recognizer in choosing acoustic models that are most appropriate for the changed conditions [7].…”

Section: Introductionmentioning

confidence: 99%

Automatic detection of high vocal effort in telephone speech

Pohjalainen¹,

Raitio²,

Pulakka³

et al. 2012

Interspeech 2012

View full text Add to dashboard Cite

A system is proposed for the automatic detection of high vocal effort in speech. The system is evaluated using both PCMcoded speech and AMR-coded telephone speech. In addition, the effect of far-end noise in the telephone conditions is studied using both matched-condition training and cases with additive noise mismatch. The proposed system is based on Bayesian classification of mel-frequency cepstral feature vectors. Concerning the MFCC feature extraction process, the substitution of a spectrum analysis method emphasizing the fine structure improves the results in the noisy cases.

show abstract

Automatic vocal effort detection for reliable speech recognition

Cited by 9 publications

References 15 publications

Comparison of methods for determining speech voicing based on tests performed on paired consonants and continuous speech

Comparison of methods for determining speech voicing based on tests performed on paired consonants and continuous speech

Speaker Discrimination Using Long-Term Spectrum of Speech

Automatic detection of high vocal effort in telephone speech

Contact Info

Product

Resources

About