Search Results for author:
Found 19 papers, 5 papers with code
Analyzing the Domain Robustness of Pretrained Language Models, Layer by Layer
The robustness of pretrained language models(PLMs) is generally measured using performance drops on two or more domains.
On the Robust Approximation of ASR Metrics
Traditionally, ASR models are evaluated using metrics like Word Error Rate (WER) and Character Error Rate (CER), which depend on ground truth labels.
Lost in Transcription, Found in Distribution Shift: Demystifying Hallucination in Speech Foundation Models
We examine how factors such as distribution shifts, model size, and model architecture influence the hallucination error rate (HER), a metric we introduce to quantify hallucinations.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+2
On the Diversity of Synthetic Data and its Impact on Training Large Language Models
The rise of Large Language Models (LLMs) has accentuated the need for diverse, high-quality pre-training data.
What Do Speech Foundation Models Not Learn About Speech?
The analysis of layer-wise features demonstrates that some models exhibit a convex relationship between the separability of the learned representations and model depth, with different layers capturing task-specific features.
uDistil-Whisper: Label-Free Data Filtering for Knowledge Distillation in Low-Data Regimes
Our models are also 25-50% more compute- and memory-efficient while maintaining performance equal to or better than that of the teacher model.
Towards Zero-Shot Text-To-Speech for Arabic Dialects
We then evaluate our models on a dataset comprising 31 unseen speakers and an in-house dialectal dataset.
To Distill or Not to Distill? On the Robustness of Robust Knowledge Distillation
Our best-distilled model's overall performance ($45. 0$\% WER) surpasses that of a SoTA model twice its size (SeamlessM4T-large-v2, WER=$47. 0$\%) and its teacher model (Whisper-large-v2, WER=$55. 1$\%), and its average performance on our new dialectal data ($56. 9$\% WER) outperforms all other models.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+3
VoxArabica: A Robust Dialect-Aware Arabic Speech Recognition System
We train a wide range of models such as HuBERT (DID), Whisper, and XLS-R (ASR) in a supervised setting for Arabic DID and ASR tasks.
A Novel Defocus-Blur Region Detection Approach Based on DCT Feature and PCNN Structure
Therefore, it is important to detect in-focused objects in defocused-blurred images after the segmentation of blurred and non-blurred regions.
TARJAMAT: Evaluation of Bard and ChatGPT on Machine Translation of Ten Arabic Varieties
Our evaluation covers diverse Arabic varieties such as Classical Arabic (CA), Modern Standard Arabic (MSA), and several country-level dialectal variants.
N-Shot Benchmarking of Whisper on Diverse Arabic Speech Recognition
Whisper, the recently developed multilingual weakly supervised model, is reported to perform well on multiple speech recognition benchmarks in both monolingual and multilingual settings.
GPTAraEval: A Comprehensive Evaluation of ChatGPT on Arabic NLP
Although we further explore and confirm the utility of employing GPT-4 as a potential alternative for human evaluation, our work adds to a growing body of research underscoring the limitations of ChatGPT.
LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions
The results demonstrate that our proposed LaMini-LM models are comparable to competitive baselines, while being much smaller in size.
Ranked #15 on
Word Sense Disambiguation
on Words in Context
Speaker and Time-aware Joint Contextual Learning for Dialogue-act Classification in Counselling Conversations
We identify the requirement of such conversation and propose twelve domain-specific dialogue-act (DAC) labels.
BloomNet: A Robust Transformer based model for Bloom's Learning Outcome Classification
For the optimization of educational programs, it is crucial to design course learning outcomes (CLOs) according to the different cognitive levels of Bloom Taxonomy.
Domain Controlled Title Generation with Human Evaluation
We study automatic title generation and present a method for generating domain-controlled titles for scientific articles.
CovidGAN: Data Augmentation Using Auxiliary Classifier GAN for Improved Covid-19 Detection
This has led to the introduction of a variety of deep learning systems and studies have shown that the accuracy of COVID-19 patient detection through the use of chest X-rays is strongly optimistic.
