Author

Feng-Ju (Claire) Chang

Applied Scientist

Filters

Applied filters

13 results found

Sort

Model-internal slot-triggered biasing for domain expansion in neural transducer ASR models

Edie Lu, Philip Harding, Kanthashree Mysore Sathyendra, Sibo Tong, Xuandi Fu, Jing Liu, Feng-Ju (Claire) Chang, Simon Wiesler, Grant Strimel

Interspeech 2023

2023

Personal rare word recognition is an important yet challenging task for end-to-end speech recognition. Contextual biasing has demonstrated success in tackling this problem. Though effective in improving rare word recognition, these mechanisms can lead to errors due to false-biasing while facing further challenges when attempting to expand them to many domains. To address these limitations, in this work

Conversational AI
Gated contextual adapters for selective contextual biasing in neural transducers

Anastasios Alexandridis, Kanthashree Mysore Sathyendra, Grant Strimel, Feng-Ju (Claire) Chang, Ariya Rastrow, Nathan Susanj, Thanasis Mouchtaris

ICASSP 2023

2023

Neural contextual biasing for end-to-end neural ASR transducers has shown significant improvements in the recognition of named entities, such as contact names or device names. However, it comes with the cost of increased compute, as the biasing layers (which are usually based on cross-attention) add complexity to the neural transducers. In this paper, we propose gated contextual biasing models that can

Conversational AI
Dialog act guided contextual adapter for personalized speech recognition

Feng-Ju (Claire) Chang, Thejaswi Muniyappa, Kanthashree Mysore Sathyendra, Kai Wei, Grant Strimel, Ross McGowan

ICASSP 2023

2023

Personalization in multi-turn dialogs has been a long standing challenge for end-to-end automatic speech recognition (E2E ASR) models. Recent work on contextual adapters has tackled rare word recognition using user catalogs. This adaptation, however, does not incorporate an important cue, the dialog act, which is available in a multi-turn dialog scenario. In this work, we propose a dialog act guided contextual

Conversational AI
Dual-attention neural transducers for efficient wake word spotting in speech recognition

Saumya Sahai, Jing Liu, Thejaswi Muniyappa, Kanthashree Mysore Sathyendra, Anastasios Alexandridis, Grant Strimel, Ross McGowan, Ariya Rastrow, Feng-Ju (Claire) Chang, Thanasis Mouchtaris, Siegfried Kunzmann

ICASSP 2023

2023

We present dual-attention neural biasing, an architecture designed to boost Wake Words (WW) recognition and improve inference time latency on speech recognition tasks. This architecture enables a dynamic switch for its runtime compute paths by exploiting WW spotting to select which branch of its attention networks to execute for an input audio frame. With this approach, we effectively improve WW spotting

Conversational AI
ConvRNN-T: Convolutional augmented recurrent neural network transducers for streaming speech recognition

Martin Radfar, Rohit Barnwal, Rupak Vignesh Swaminathan, Feng-Ju (Claire) Chang, Grant Strimel, Nathan Susanj, Thanasis Mouchtaris

Interspeech 2022

2022

The recurrent neural network transducer (RNN-T) is a prominent streaming end-to-end (E2E) ASR technology. In RNN-T, the acoustic encoder commonly consists of stacks of LSTMs. Very recently, as an alternative to LSTM layers, the Conformer architecture was introduced where the encoder of RNN-T is replaced with a modified Transformer encoder composed of convolutional layers at the frontend and between attention

Related: Alexa speech science developments at Interspeech 2022

Conversational AI
Compute cost amortized transformer for streaming ASR

Yi Xie, Jonathan Macoskey, Martin Radfar, Feng-Ju (Claire) Chang, Brian King, Ariya Rastrow, Thanasis Mouchtaris, Grant Strimel

Interspeech 2022

2022

We present a streaming, Transformer-based end-to-end automatic speech recognition (ASR) architecture which achieves efficient neural inference through compute cost amortization. Our architecture creates sparse computation pathways dynamically at inference time, resulting in selective use of compute resources throughout decoding, enabling significant reductions in compute with minimal impact on accuracy.

Related: Alexa speech science developments at Interspeech 2022

Conversational AI
Contextual adapters for personalized speech recognition in neural transducers

Kanthashree Mysore Sathyendra, Thejaswi Muniyappa, Feng-Ju (Claire) Chang, Jing Liu, Jinru Su, Grant Strimel, Thanasis Mouchtaris, Siegfried Kunzmann

ICASSP 2022

2022

Personal rare word recognition in end-to-end Automatic Speech Recognition (E2E ASR) models is a challenge due to the lack of training data. A standard way to address this issue is with shallow fusion methods at inference time. However, due to their dependence on external language models and the deterministic approach to weight boosting, their performance is limited. In this paper, we propose training neural

Conversational AI
Multi-task RNN-T with semantic decoder for streamable spoken language understanding

Xuandi Fu, Feng-Ju (Claire) Chang, Martin Radfar, Kai Wei, Jing Liu, Grant Strimel, Kanthashree Mysore Sathyendra

ICASSP 2022

2022

End-to-end Spoken Language Understanding (E2E SLU) has attracted increasing interest due to its advantages of joint optimization and low latency when compared to traditionally cascaded pipelines. Existing E2E SLU models usually follow a two-stage configuration where an Automatic Speech Recognition (ASR) network first predicts a transcript which is then passed to a Natural Language Understanding (NLU) module

Conversational AI
Attentive contextual carryover for multi-turn end-to-end spoken language understanding

Kai Wei, Thanh Tran, Feng-Ju (Claire) Chang, Kanthashree Mysore Sathyendra, Thejaswi Muniyappa, Jing Liu, Anirudh Raju, Ross McGowan, Nathan Susanj, Ariya Rastrow, Grant Strimel

ASRU 2021

2021

Recent years have seen significant advances in end-to-end (E2E) spoken language understanding (SLU) systems, which directly predict intents and slots from spoken audio. While dialogue history has been exploited to improve conventional text-based natural language understanding systems, current E2E SLU approaches have not yet incorporated such critical contextual signals in multi-turn and task-oriented dialogues

Conversational AI
Context-aware Transformer transducer for speech recognition

Feng-Ju (Claire) Chang, Jing Liu, Martin Radfar, Thanasis Mouchtaris, Maurizio Omologo, Ariya Rastrow, Siegfried Kunzmann

ASRU 2021

2021

End-to-end (E2E) automatic speech recognition (ASR) systems often have difficulty recognizing uncommon words, that appear infrequently in the training data. One promising method, to improve the recognition accuracy on such rare words, is to latch onto personalized/contextual information at inference. In this work, we present a novel context-aware transformer transducer (CATT) network that improves the state-of-the-art

Related: ASRU: Integrating speech recognition and language understanding

Conversational AI