Feature learning for efficient ASR-free keyword spotting in low-resource languages
We consider feature learning for a computationally efficient method of keyword spotting that
can be applied in severely under-resourced settings. The objective is to support …
can be applied in severely under-resourced settings. The objective is to support …
Feature exploration for almost zero-resource ASR-free keyword spotting using a multilingual bottleneck extractor and correspondence autoencoders
We compare features for dynamic time warping (DTW) when used to bootstrap keyword
spotting (KWS) in an almost zero-resource setting. Such quickly-deployable systems aim to …
spotting (KWS) in an almost zero-resource setting. Such quickly-deployable systems aim to …
The makerere radio speech corpus: A Luganda radio corpus for automatic speech recognition
J Mukiibi, A Katumba, J Nakatumba-Nabende… - arXiv preprint arXiv …, 2022 - arxiv.org
Building a usable radio monitoring automatic speech recognition (ASR) system is a
challenging task for under-resourced languages and yet this is paramount in societies …
challenging task for under-resourced languages and yet this is paramount in societies …
Towards hate speech detection in low-resource languages: Comparing ASR to acoustic word embeddings on Wolof and Swahili
We consider hate speech detection through keyword spotting on radio broadcasts. One
approach is to build an automatic speech recognition (ASR) system for the target low …
approach is to build an automatic speech recognition (ASR) system for the target low …
Fast ASR-free and almost zero-resource keyword spotting using DTW and CNNs for humanitarian monitoring
We use dynamic time warping (DTW) as supervision for training a convolutional neural
network (CNN) based keyword spotting system using a small set of spoken isolated …
network (CNN) based keyword spotting system using a small set of spoken isolated …
[PDF][PDF] Multilingual Neural Network Acoustic Modelling for ASR of Under-Resourced English-isiZulu Code-Switched Speech.
Although isiZulu speakers code-switch with English as a matter of course, extremely little
appropriate data is available for acoustic modelling. Recently, a small five-language corpus …
appropriate data is available for acoustic modelling. Recently, a small five-language corpus …
Improved low-resource Somali speech recognition by semi-supervised acoustic and language model training
We present improvements in automatic speech recognition (ASR) for Somali, a currently
extremely under-resourced language. This forms part of a continuing United Nations (UN) …
extremely under-resourced language. This forms part of a continuing United Nations (UN) …
[PDF][PDF] Improving ASR for Code-Switched Speech in Under-Resourced Languages Using Out-of-Domain Data.
We explore the use of out-of-domain monolingual data for the improvement of automatic
speech recognition (ASR) of codeswitched speech. This is relevant because annotated …
speech recognition (ASR) of codeswitched speech. This is relevant because annotated …
Multilingual acoustic word embeddings for zero-resource languages
C Jacobs - arXiv preprint arXiv:2401.10543, 2024 - arxiv.org
This research addresses the challenge of developing speech applications for zero-resource
languages that lack labelled data. It specifically uses acoustic word embedding (AWE)--fixed …
languages that lack labelled data. It specifically uses acoustic word embedding (AWE)--fixed …
ASR-free CNN-DTW keyword spotting using multilingual bottleneck features for almost zero-resource languages
We consider multilingual bottleneck features (BNFs) for nearly zero-resource keyword
spotting. This forms part of a United Nations effort using keyword spotting to support …
spotting. This forms part of a United Nations effort using keyword spotting to support …