In air traffic control (ATC), speech communication with radio transmission is the primary way
to exchange information between the controller and aircrew. A wealth of contextual …

In this work, we propose a new parameter-efficient learning framework based on neural
model reprogramming for cross-lingual speech recognition, which can re-purpose well …

In recent years, all-neural, end-to-end (E2E) ASR systems gained rapid interest in the
speech recognition community. They convert speech input to text units in a single trainable …

Low-resource automatic speech recognition (ASR) is challenging, as the low-resource target
language data cannot well train an ASR model. To solve this issue, meta-learning …

In this paper, a multilingual end-to-end framework, called ATCSpeechNet, is proposed to
tackle the issue of translating communication speech into human-readable text in air traffic …

Automatic speech recognition (ASR) has been significantly advanced with the use of deep
learning and big data. How-ever improving robustness, including achieving equally good …

Hotword customization is one of the concerned issues remained in ASR field-it is of value to
enable users of ASR systems to customize names of entities, persons and other phrases to …

Building language-universal speech recognition systems entails producing phonological
units of spoken sound that can be shared across languages. While speech annotations at …

Contextual biasing enables speech recognizers to transcribe important phrases in the
speaker's context, such as contact names, even if they are rare in, or absent from, the …

Following the rationale of end-to-end modeling, CTC, RNN-T or encoder-decoder-attention
models for automatic speech recognition (ASR) use graphemes or grapheme-based …