Cross-attention conformer for context modeling in speech enhancement for ASR
This work addresses noise robustness in ASR systems, which is a domain-specific incremental improvement.
The paper tackles the problem of improving noise robustness in automatic speech recognition by using a cross-attention conformer architecture to model contextual information, such as noise-only audio segments, for speech enhancement, resulting in enhanced features for ASR.
This work introduces \emph{cross-attention conformer}, an attention-based architecture for context modeling in speech enhancement. Given that the context information can often be sequential, and of different length as the audio that is to be enhanced, we make use of cross-attention to summarize and merge contextual information with input features. Building upon the recently proposed conformer model that uses self attention layers as building blocks, the proposed cross-attention conformer can be used to build deep contextual models. As a concrete example, we show how noise context, i.e., short noise-only audio segment preceding an utterance, can be used to build a speech enhancement feature frontend using cross-attention conformer layers for improving noise robustness of automatic speech recognition.