Investigations on End-to-End Audiovisual Fusion
This work addresses noise robustness in audiovisual speech recognition, but it is incremental as it builds on existing deep learning methods without a major breakthrough.
The paper tackles the problem of noise in speech recognition by proposing an end-to-end audiovisual fusion neural network, which outperforms single-modality recognition across all noise conditions.
Audiovisual speech recognition (AVSR) is a method to alleviate the adverse effect of noise in the acoustic signal. Leveraging recent developments in deep neural network-based speech recognition, we present an AVSR neural network architecture which is trained end-to-end, without the need to separately model the process of decision fusion as in conventional (e.g. HMM-based) systems. The fusion system outperforms single-modality recognition under all noise conditions. Investigation of the saliency of the input features shows that the neural network automatically adapts to different noise levels in the acoustic signal.