English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Audiovisual automatic speech recognition (AV-ASR) aims to improve therobustness of a speech recognition system by incorporating visual information.Training fully supervised multimodal models for this task from scratch, howeveris limited by the need for large labelled audiovisual datasets (in eachdownstream domain of interest). We present AVFormer, a simple method foraugmenting audio-only models with visual information, at the same timeperforming lightweight domain adaptation. We do this by (i) injecting visualembeddings into a frozen ASR model using lightweight trainable adaptors. Weshow that these can be trained on a small amount of weakly labelled video datawith minimum additional training time and parameters. (ii) We also introduce asimple curriculum scheme during training which we show is crucial to enable themodel to jointly process audio and visual information effectively; and finally(iii) we show that our model achieves state of the art zero-shot results onthree different AV-ASR benchmarks (How2, VisSpeech and Ego4D), while alsocrucially preserving decent performance on traditional audio-only speechrecognition benchmarks (LibriSpeech). Qualitative results show that our modeleffectively leverages visual information for robust speech recognition.

AVFormer: Injecting Vision into Frozen Speech Models for Zero-Shot AV-ASR