End-to-end Continuous Speech Recognition using Attention-based Recurrent NN: First Results
This addresses speech recognition for applications requiring end-to-end processing, but it is incremental as it matches rather than surpasses existing methods.
The paper tackled continuous speech recognition by replacing traditional HMMs with an attention-based recurrent neural network that directly outputs phonemes, achieving phoneme error rates comparable to state-of-the-art HMM-based decoders on the TIMIT dataset.
We replace the Hidden Markov Model (HMM) which is traditionally used in in continuous speech recognition with a bi-directional recurrent neural network encoder coupled to a recurrent neural network decoder that directly emits a stream of phonemes. The alignment between the input and output sequences is established using an attention mechanism: the decoder emits each symbol based on a context created with a subset of input symbols elected by the attention mechanism. We report initial results demonstrating that this new approach achieves phoneme error rates that are comparable to the state-of-the-art HMM-based decoders, on the TIMIT dataset.