… We explored very deep CNNs for end-to-end speech recognition. We applied Network-in-Network
principles to add depth and nonlinearities to hierarchical RNNs. We also applied …

… Recently, there has been a strong push to transition from hybrid models to end-to-end (E2E)
models for automatic speech recognition. Currently, there are three promising E2E methods: …

… We design an online end-to-end speech recognition system based on Time-Depth Separable
(TDS) convolutions and Connectionist Temporal Classification (CTC). We improve the …

… To the best of our knowledge, all end-to-end neural speech recognition systems employ
recurrent neural networks in at least some part of the processing pipeline. The most successful …

… In this work, we present a novel, all-neural, end-to-end (E2E) ASR system that utilizes such
context. Our approach, which we refer to as Contextual Listen, Attend and Spell (CLAS) jointly…

… Deep speech 2: End-to-end speech recognition in english and mandarin. In Proceedings
of the International Conference on Machine Learning, New York, NY, USA, 19–24 June 2016; …

… , and we show that a competitive end-to-end ASR model can be achieved solely using …
end-to-end ASR models with the Transformer. Second, in order to facilitate training of very deep …

… Although their implementation was followed as closely as possible, training end-toend
quickly exceeded the memory limitations of modern GPUs. To work around these problems, the …

… on challenging tasks such as voice search. In … end-to-end (E2E) models [10, 11, 12, 13,
14]. Such models replace the traditional components of an ASR system with a single, end-to-end …

… Recently, the speech community is seeing a significant trend of moving from deep neural
network based hybrid modeling to end-to-end (E2E) modeling for automatic speech recognition …