It is commonly accepted that morphological dependencies are finite-state in nature. We argue that the upper bound on morphological expressivity is much lower. Drawing on technical results from computational phonology, we show that a variety of morphotactic phenomena are tierbased strictly local and do not fall into weaker subclasses such as the strictly local or strictly piecewise languages. Since the tier-based strictly local languages are learnable in the limit from positive texts, this marks a first important step towards general machine learning algorithms for morphology. Furthermore, the limitation to tier-based strictly local languages explains typological gaps that are puzzling from a purely linguistic perspective.

The applications of automatic speech recognition (ASR) systems are proliferating, in part due to recent significant quality improvements. However, as recent work indicates, even state-of-the-art speech recognition systems -some which deliver impressive benchmark results, struggle to generalize across use cases. We review relevant work, and, hoping to inform future benchmark development, outline a taxonomy of speech recognition use cases, proposed for the next generation of ASR benchmarks. We also survey work on metrics, in addition to the de facto standard Word Error Rate (WER) metric, and we introduce a versatile framework designed to describe interactions between linguistic variation and ASR performance metrics.

No abstract

No abstract

Building inclusive speech recognition systems is a crucial step towards developing technologies that speakers of all language varieties can use. Therefore, ASR systems must work for everybody independently of the way they speak. To accomplish this goal, there should be available data sets representing language varieties, and also an understanding of model configuration that is the most helpful in achieving robust understanding of all types of speech. However, there are not enough data sets for accented speech, and for the ones that are already available, more training approaches need to be explored to improve the quality of accented speech recognition. In this paper, we discuss recent progress towards developing more inclusive ASR systems, namely, the importance of building new data sets representing linguistic diversity, and exploring novel training approaches to improve performance for all users. We address recent directions within benchmarking ASR systems for accented speech, measure the effects of wav2vec 2.0 pre-training on accented speech recognition, and highlight corpora relevant for diverse ASR evaluations.