Auditory Features Revisited for Robust Speech Recognition

Kelly, Finnian; Harte, Naomi

doi:10.1109/icpr.2010.1082

Cited by 8 publications

(2 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In rows (1) and (2) Caranica et al [47] used supervised phoneme HMM tokens trained on a labeled Romanian corpus of 8.7 hours. The results of row (1) were trained with MFCCs, and row (2) with PNCCs [53]. Their systems are similar to the proposed approach here because we both use token HMMs.…”

Section: Spoken Term Detection Experimentsmentioning

confidence: 99%

Unsupervised Discovery of Structured Acoustic Tokens With Applications to Spoken Term Detection

Chung

Lee

2018

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

In this paper, we compare two paradigms for unsupervised discovery of structured acoustic tokens directly from speech corpora without any human annotation. The Multigranular Paradigm seeks to capture all available information in the corpora with multiple sets of tokens for different model granularities. The Hierarchical Paradigm attempts to jointly learn several levels of signal representations in a hierarchical structure. The two paradigms are unified within a theoretical framework in this paper. Query-by-Example Spoken Term Detection (QbE-STD) experiments on the QUESST dataset of MediaEval 2015 verifies the competitiveness of the acoustic tokens. The Enhanced Relevance Score (ERS) proposed in this work improves both paradigms for the task of QbE-STD. We also list results on the ABX evaluation task of the Zero Resource Challenge 2015 for comparison of the Paradigms.

show abstract

Section: Spoken Term Detection Experimentsmentioning

confidence: 99%

Unsupervised Discovery of Structured Acoustic Tokens With Applications to Spoken Term Detection

Chung

Lee

2018

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

show abstract

“…The resulting feature set has been shown to lead to a better recognition accuracy than MFCC or RASTA-PLP processing in the presence of common additive noise types, and reverberation [113]. It was also heroded as the new feature set that is the most promising for robust ASR [108,109]. However the noise robustness of the PNCC has a downside because it has a lower performance (compared to MFCC or PLP) under clean speech conditions [168].…”

Section: Power Normalised Cepstral Coefficients (Pncc)mentioning

confidence: 99%