We find that, although maximising the training pool by including all six additional languages provides improved speech recognition in both target languages, substantially better performance can be achieved by a more judicious choice.
We show that, using these features, the CNN-DTW keyword spotter performs almost as well as the DTW keyword spotter while outperforming a baseline CNN trained only on the keyword templates.
In this work, we explore the benefits of using multilingual bottleneck features (mBNF) in acoustic modelling for the automatic speech recognition of code-switched (CS) speech in African languages.
We present an analysis of semi-supervised acoustic and language model training for English-isiZulu code-switched (CS) ASR using soap opera speech.
Automatic segmentation was applied in combination with automaticspeaker diarization.
This paper reports on the semi-supervised development of acoustic and language models for under-resourced, code-switched speech in five South African languages.
The automatic transcriptions from the best performing pass were used for language model augmentation.
Furthermore, because English is common to all language pairs in our data, it dominates when training a unified language model, leading to improved English ASR performance at the expense of the other languages.
no code implementations • 16 Apr 2019 • Ryan Eloff, André Nortje, Benjamin van Niekerk, Avashna Govender, Leanne Nortje, Arnu Pretorius, Elan van Biljon, Ewald van der Westhuizen, Lisa van Staden, Herman Kamper
For our submission to the ZeroSpeech 2019 challenge, we apply discrete latent-variable neural networks to unlabelled speech and use the discovered units for speech synthesis.
We compare features for dynamic time warping (DTW) when used to bootstrap keyword spotting (KWS) in an almost zero-resource setting.
We present our first efforts towards building a single multilingual automatic speech recognition (ASR) system that can process code-switching (CS) speech in five languages spoken within the same population.