5 papers
Normal-Anchored First-Order Model-Agnostic Meta-Learning based Whisper Fine-Tuning for Enhancing Fairness of Cleft Lip and Palate Speech Recognition
Susmita Bhattacharjee, Jagabandhu Mishra, H. S. Shekhawat +2
Automatic speech recognition (ASR) for cleft lip and palate (CLP) speech is difficult because acoustic and articulatory patterns vary across severity levels. This variability reduc…
Tensor-based empirical interpolation method and its application in model reduction
Brij Nandan Tripathi, Hanumant Singh Shekhawat, Seip Weiland
In general, matrix or tensor-valued functions are approximated using the method developed for vector-valued functions by transforming the matrix-valued function into vector form. T…
Parameter-Efficient Fine-Tuning of Foundation Models for CLP Speech Classification
Susmita Bhattacharjee, Jagabandhu Mishra, H. S. Shekhawat +1
We propose the use of parameter-efficient fine-tuning (PEFT) of foundation models for cleft lip and palate (CLP) detection and severity classification. In CLP, nasalization increas…
Improving ASR Fairness for Cleft Lip and Palate Speech: A Study on Severity-Aware Data Mixing
Susmita Bhattacharjee, Jagabandhu Mishra, H. S. Shekhawat +2
Speech produced by individuals with cleft lip and palate (CLP) is often hypernasal (and sometimes breathy) due to structural anomalies, yielding shifts in formant structure that de…
Enhanced Error Bounds For The Masked Projection Techniques via Cosine-Sine Decomposition
Brij Nandan Tripathi, Hanumant Singh Shekhawat
The masked projection techniques are popular in the area of non-linear model reduction. Quantifying and minimizing the error in model reduction, particularly from masked projection…