[PDF][PDF] A review of speech-centric trustworthy machine learning: Privacy, safety, and fairness
Speech-centric machine learning systems have revolutionized a number of leading
industries ranging from transportation and healthcare to education and defense …
industries ranging from transportation and healthcare to education and defense …
Deep representation learning in speech processing: Challenges, recent advances, and future trends
Research on speech processing has traditionally considered the task of designing hand-
engineered acoustic features (feature engineering) as a separate distinct problem from the …
engineered acoustic features (feature engineering) as a separate distinct problem from the …
Massively multilingual adversarial speech recognition
We report on adaptation of multilingual end-to-end speech recognition models trained on as
many as 100 languages. Our findings shed light on the relative importance of similarity …
many as 100 languages. Our findings shed light on the relative importance of similarity …
Universlu: Universal spoken language understanding for diverse classification and sequence generation tasks with a single network
Recent studies have demonstrated promising outcomes by employing large language
models with multi-tasking capabilities. They utilize prompts to guide the model's behavior …
models with multi-tasking capabilities. They utilize prompts to guide the model's behavior …
End-to-end domain-adversarial voice activity detection
Voice activity detection is the task of detecting speech regions in a given audio stream or
recording. First, we design a neural network combining trainable filters and recurrent layers …
recording. First, we design a neural network combining trainable filters and recurrent layers …
A study of bias mitigation strategies for speaker recognition
Speaker recognition is increasingly used in several everyday applications including smart
speakers, customer care centers and other speech-driven analytics. It is crucial to accurately …
speakers, customer care centers and other speech-driven analytics. It is crucial to accurately …
Best of both worlds: Robust accented speech recognition with adversarial transfer learning
Training deep neural networks for automatic speech recognition (ASR) requires large
amounts of transcribed speech. This becomes a bottleneck for training robust models for …
amounts of transcribed speech. This becomes a bottleneck for training robust models for …
Unsupervised domain adaptation schemes for building ASR in low-resource languages
CS Anoop, AP Prathosh… - 2021 IEEE Automatic …, 2021 - ieeexplore.ieee.org
Building an automatic speech recognition (ASR) system from scratch requires a large
amount of annotated speech data, which is difficult to collect in many languages. However …
amount of annotated speech data, which is difficult to collect in many languages. However …
Adversarial regularization for attention based end-to-end robust speech recognition
End-to-end speech recognition, such as attention based approaches, is an emerging and
attractive topic in recent years. It has achieved comparable performance with the traditional …
attractive topic in recent years. It has achieved comparable performance with the traditional …
Disentangled speaker and nuisance attribute embedding for robust speaker verification
Over the recent years, various deep learning-based embedding methods have been
proposed and have shown impressive performance in speaker verification. However, as in …
proposed and have shown impressive performance in speaker verification. However, as in …