Statistical normalisation of phase-based feature representation for robust speech recognition

Loweimi, Erfan; Barker, Jon; Hain, Thomas

doi:10.1109/icassp.2017.7953170

Cited by 11 publications

(6 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…That is, the quality/intelligibility of the magnitude-only reconstructed signal is significantly less than the original one. This point has been verified by both subjective [38], [39] and objective [28], [40]- [43] tests. Such detrimental information loss which is important from both theoretical and perceptual perspectives, potentially harms the performance of the raw magnitude-based systems.…”

Section: Information Content Of the Ft's Componentsmentioning

confidence: 59%

Multi-Stream Acoustic Modelling Using Raw Real and Imaginary Parts of the Fourier Transform

Loweimi

Yue²,

Bell

et al. 2023

IEEE/ACM Trans. Audio Speech Lang. Process.

Self Cite

View full text Add to dashboard Cite

Section: Information Content Of the Ft's Componentsmentioning

confidence: 59%

Multi-Stream Acoustic Modelling Using Raw Real and Imaginary Parts of the Fourier Transform

Loweimi

Yue²,

Bell

et al. 2023

IEEE/ACM Trans. Audio Speech Lang. Process.

Self Cite

View full text Add to dashboard Cite

“…GFNet extracts frequency domain features by multiplying the real and imaginary parts with same parameters, which maintains all phase spectrum. However, the quality of the signal reconstructed using only the amplitude is noticeably lower than that of the original signal (Loweimi et al 2017). The phase spectrum is solely related to the real part and the imaginary part of each frequency.…”

Section: Fda Blockmentioning

confidence: 95%

Sacroiliitis diagnosis based on interpretable features and multi-task learning

Liu,

Zhang,

Zhang

et al. 2024

Phys. Med. Biol.

View full text Add to dashboard Cite

Objective. Sacroiliitis is an early pathological manifestation of ankylosing spondylitis (AS), and a positive sacroiliitis test on imaging may help clinical practitioners diagnose AS early. Deep learning based automatic diagnosis algorithms can deliver grading findings for sacroiliitis, however, it requires a large amount of data with precise labels to train the model and lacks grading features visualization. In this paper, we aimed to propose a radiomics and deep learning based deep feature visualization positive diagnosis algorithm for sacroiliitis on CT scans. Visualization of grading features can enhance clinical interpretability with visual grading features, which assist doctors in diagnosis and treatment more effectively.Approach. The region of interest (ROI) is identified by segmenting the sacroiliac joint (SIJ) 3D CT images using a combination of the U-net model and certain statistical approaches. Then, in addition to extracting spatial and frequency domain features from ROI according to the radiographic manifestations of sacroiliitis, the radiomics features have also been integrated into the proposed encoder module to obtain a powerful encoder and extract features effectively. Finally, a multi-task learning technique and five-class labels are utilized to help with performing positive tests to reduce discrepancies in the evaluation of several radiologists.Main results. On our private dataset, proposed methods have obtained an accuracy rate of 87.3$\%$, which is 9.8$\%$ higher than the baseline and consistent with assessments made by qualified medical professionals.Significance. The results of the ablation experiment and interpreting analysis demonstrated that the proposed methods are applied in automatic CT scan sacroiliitis diagnosis due to their excellently interpretable and portable advantages.

show abstract

“…In [19], the statistical properties of the phase spectrum and phase-based representations were scrutinised. Moreover, usefulness of various types of statistical normalisation methods (histogram equalisation, Gaussianisation and mean-variance normalisation) of the phase-based features at different stages along the pipeline were investigated on the A2 task.…”

Section: Phase Spectrum Applications In Asrmentioning

confidence: 99%

Speech Acoustic Modelling from Raw Phase Spectrum

Loweimi

Cvetković

Bell

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

Magnitude spectrum-based features are the most widely employed front-ends for acoustic modelling in automatic speech recognition (ASR) systems. In this paper, we investigate the possibility and efficacy of acoustic modelling using the raw short-time phase spectrum. In particular, we study the usefulness of the raw wrapped, unwrapped and minimum-phase phase spectra as well as the phase of the source and filter components for acoustic modelling. Furthermore, we explore the effectiveness of simultaneous deployment of the vocal tract and excitation components of the raw phase spectrum using multi-head CNNs and investigate multiple information fusion schemes. This paves the way for developing an effective phase-based multi-stream information processing systems for speech recognition. The performance, even for wrapped phase with a noise-like shape, is comparable to or better than the magnitude-based classic features, and up to 4.8% WER has been achieved in the WSJ (Eval-92) task.

show abstract

Statistical normalisation of phase-based feature representation for robust speech recognition

Cited by 11 publications

References 20 publications

Multi-Stream Acoustic Modelling Using Raw Real and Imaginary Parts of the Fourier Transform

Multi-Stream Acoustic Modelling Using Raw Real and Imaginary Parts of the Fourier Transform

Sacroiliitis diagnosis based on interpretable features and multi-task learning

Speech Acoustic Modelling from Raw Phase Spectrum

Contact Info

Product

Resources

About