CIF: Continuous Integrate-and-Fire for End-to-End Speech Recognition
arXiv:1905.11235
Abstract
In this paper, we propose a novel soft and monotonic alignment mechanism used for sequence transduction. It is inspired by the integrate-and-fire model in spiking neural networks and employed in the encoder-decoder framework consists of continuous functions, thus being named as: Continuous Integrate-and-Fire (CIF). Applied to the ASR task, CIF not only shows a concise calculation, but also supports online recognition and acoustic boundary positioning, thus suitable for various ASR scenarios. Several support strategies are also proposed to alleviate the unique problems of CIF-based model. With the joint action of these methods, the CIF-based model shows competitive performance. Notably, it achieves a word error rate (WER) of 2.86% on the test-clean of Librispeech and creates new state-of-the-art result on Mandarin telephone ASR benchmark.
arXiv commentTo appear at ICASSP 2020
References in corpus (14)
- SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
- Attention-Based Models for Speech Recognition
- Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks
- Sequence Transduction with Recurrent Neural Networks
- Layer Normalization
- Improved training of end-to-end attention models for speech recognition
- AISHELL-2: Transforming Mandarin ASR Research Into Industrial Scale
- wav2letter++: The Fastest Open-source Speech Recognition System
- Transformers with convolutional context for ASR
- Self-Attention Networks for Connectionist Temporal Classification in Speech Recognition
- ESPnet: End-to-End Speech Processing Toolkit
- Local Monotonic Attention Mechanism for End-to-End Speech and Language Processing
- Letter-Based Speech Recognition with Gated ConvNets
- Monotonic Chunkwise Attention
Cited by in corpus (4)
- Gated Recurrent Context: Softmax-free Attention for Online Encoder-Decoder Speech Recognition
- FastLR: Non-Autoregressive Lipreading Model with Integrate-and-Fire
- Minimum Latency Training Strategies for Streaming Sequence-to-Sequence ASR
- Towards efficient end-to-end speech recognition with biologically-inspired neural networks