-
ICASSP 20212021Transformers are powerful neural architectures that allow integrating different modalities using attention mechanisms. In this paper, we leverage the neural transformer architectures for multi-channel speech recognition systems, where the spectral and spatial information collected from different microphones are integrated using attention layers. Our multi-channel transformer network mainly consists of three
-
ICASSP 20212021Automatic dubbing aims at seamlessly replacing the speech in a video document with synthetic speech in a different language. The task implies many challenges, one of which is generating translations that not only convey the original content, but also match the duration of the corresponding utterances. In this paper, we focus on the problem of controlling the verbosity of machine translation output, so that
-
ICASSP 20212021Automatic dubbing is an extension of speech-to-speech translation such that the resulting target speech is carefully aligned in terms of duration, lip movements, timbre, emotion, prosody, etc. of the speaker in order to achieve audiovisual coherence. Dubbing quality strongly depends on isochrony, i.e., arranging the translation of the original speech to optimally match its sequence of phrases and pauses
-
ICASSP 20212021Accents mismatching is a critical problem for end-to-end ASR. This paper aims to address this problem by building an accent-robust RNN-T system with domain adversarial training (DAT). We unveil the magic behind DAT and provide, for the first time, a theoretical guarantee that DAT learns accentinvariant representations. We also prove that performing the gradient reversal in DAT is equivalent to minimizing
-
ICASSP 20212021We present Bifocal RNN-T, a new variant of the Recurrent Neural Network Transducer (RNN-T) architecture designed for improved inference time latency on speech recognition tasks. The architecture enables a dynamic pivot for its runtime compute pathway, namely taking advantage of keyword spotting to select which component of the network to execute for a given audio frame. To accomplish this, we leverage a
Related content
-
June 08, 2022New method would enable BERT-based natural-language-processing models to handle longer text strings, run in resource-constrained settings — or sometimes both. -
Based on a figure from "TernaryBERT: Distillation-aware ultra-low bit BERT"June 06, 2022Combination of distillation and distillation-aware quantization compresses BART model to 1/16th its size.
-
June 01, 2022Knowledge distillation and discriminative training enable efficient use of a BERT-based model to rescore automatic-speech-recognition hypotheses.
-
May 27, 2022Amazon Scholar and Columbia professor Kathleen McKeown on model compression, data distribution shifts, language revitalization, and more.
-
May 17, 2022Papers focus on speech conversion and data augmentation — and sometimes both at once.
-
May 12, 2022Multimodal training, signal-to-interpretation, and BERT rescoring are just a few topics covered by Amazon’s 21 speech-related papers.
