Sequence Transduction with Recurrent Neural Networks

Abstract

Many machine learning tasks can be expressed as the transformation---or \emph{transduction}---of input sequences into output sequences: speech recognition, machine translation, protein secondary structure prediction and text-to-speech to name but a few. One of the key challenges in sequence transduction is learning to represent both the input and output sequences in a way that is invariant to sequential distortions such as shrinking, stretching and translating. Recurrent neural networks (RNNs) are a powerful sequence learning architecture that has proven capable of learning such representations. However RNNs traditionally require a pre-defined alignment between the input and output sequences to perform transduction. This is a severe limitation since \emph{finding} the alignment is the most difficult aspect of many sequence transduction problems. Indeed, even determining the length of the output sequence is often challenging. This paper introduces an end-to-end, probabilistic sequence transduction system, based entirely on RNNs, that is in principle able to transform any input sequence into any finite, discrete output sequence. Experimental results for phoneme recognition are provided on the TIMIT speech corpus.

First published in the International Conference of Machine Learning (ICML) 2012 Workshop on Representation Learning

Cited by in corpus (392)

• Neural Machine Translation by Jointly Learning to Align and Translate
• Attention-Based Models for Speech Recognition
• On the Properties of Neural Machine Translation: Encoder-Decoder Approaches
• Deep Learning for Audio Signal Processing
• Deep Audio-Visual Speech Recognition
• SpeechBrain: A General-Purpose Speech Toolkit
• End-to-end Continuous Speech Recognition using Attention-based Recurrent NN: First Results
• Conformer: Convolution-augmented Transformer for Speech Recognition
• Listen, Attend and Spell
• An overview and comparative analysis of Recurrent Neural Networks for Short Term Load Forecasting
• Improved Noisy Student Training for Automatic Speech Recognition
• Conditional molecular design with deep generative models
• Predicting Domain Generation Algorithms with Long Short-Term Memory Networks
• Pushing the Limits of Semi-Supervised Learning for Automatic Speech Recognition
• A Survey of Available Corpora for Building Data-Driven Dialogue Systems
• BigSSL: Exploring the Frontier of Large-Scale Semi-Supervised Learning for Automatic Speech Recognition
• Learning to Transduce with Unbounded Memory
• Distributed Compressive Sensing: A Deep Learning Approach
• Continual Learning for Recurrent Neural Networks: an Empirical Evaluation
• Curriculum Learning and Minibatch Bucketing in Neural Machine Translation
• Transforming Question Answering Datasets Into Natural Language Inference Datasets
• Personalizing ASR for Dysarthric and Accented Speech with Limited Data
• A Survey of Deep Learning Techniques for Neural Machine Translation
• Recent Progress in the CUHK Dysarthric Speech Recognition System
• Self-Attention Transducers for End-to-End Speech Recognition
• Adaptation Algorithms for Neural Network-Based Speech Recognition: An Overview
• SpeechStew: Simply Mix All Available Speech Recognition Data to Train One Large Neural Network
• Training Speech Recognition Models with Federated Learning: A Quality/Cost Framework
• Guiding Long-Short Term Memory for Image Caption Generation
• Exploring Neural Transducers for End-to-End Speech Recognition
• ContextNet: Improving Convolutional Neural Networks for Automatic Speech Recognition with Global Context
• Towards End-to-End Speech Recognition with Deep Convolutional Neural Networks
• Transformer-Transducer: End-to-End Speech Recognition with Self-Attention
• EdgeDRNN: Recurrent Neural Network Accelerator for Edge Inference
• Applying Wav2vec2.0 to Speech Recognition in Various Low-resource Languages
• Optimizing Speech Recognition For The Edge
• Pervasive Attention: 2D Convolutional Neural Networks for Sequence-to-Sequence Prediction
• A Survey on State-of-the-art Deep Learning Applications and Challenges
• Unified Streaming and Non-streaming Two-pass End-to-end Model for Speech Recognition
• Automated Audio Captioning: An Overview of Recent Progress and New Challenges
• Text Generation from Knowledge Graphs with Graph Transformers
• Citrinet: Closing the Gap between Non-Autoregressive and Autoregressive End-to-End Models for Automatic Speech Recognition
• Streaming end-to-end multi-talker speech recognition
• A New Training Pipeline for an Improved Neural Transducer
• Speaker Adaptation for Attention-Based End-to-End Speech Recognition
• UniSpeech: Unified Speech Representation Learning with Labeled and Unlabeled Data
• On the Choice of Modeling Unit for Sequence-to-Sequence Speech Recognition
• Implicit Distortion and Fertility Models for Attention-based Encoder-Decoder NMT Model
• Speech To Semantics: Improve ASR and NLU Jointly via All-Neural Interfaces
• Sequence Modeling via Segmentations
• A Neural Transducer
• TACO: Learning Task Decomposition via Temporal Alignment for Control
• State-of-the-art Speech Recognition using EEG and Towards Decoding of Speech Spectrum From EEG
• RNN-T For Latency Controlled ASR With Improved Beam Search
• Technical Q&A Site Answer Recommendation via Question Boosting
• End-to-End Adversarial Text-to-Speech
• Monotonic Chunkwise Attention
• A CTC Alignment-based Non-autoregressive Transformer for End-to-end Automatic Speech Recognition
• Towards Online End-to-end Transformer Automatic Speech Recognition
• Improving Variational Encoder-Decoders in Dialogue Generation
• Deep Recurrent Neural Networks for Acoustic Modelling
• Transformer Transducer: One Model Unifying Streaming and Non-streaming Speech Recognition
• End-to-End Neural Segmental Models for Speech Recognition
• A Conditional Variational Framework for Dialog Generation
• Non-autoregressive Transformer-based End-to-end ASR using BERT
• Tied & Reduced RNN-T Decoder
• Transformer Transducer: A Streamable Speech Recognition Model with Transformer Encoders and RNN-T Loss
• From English to More Languages: Parameter-Efficient Model Reprogramming for Cross-Lingual Speech Recognition
• Dual-mode ASR: Unify and Improve Streaming ASR with Full-context Modeling
• TSNAT: Two-Step Non-Autoregressvie Transformer Models for Speech Recognition
• Semantic Mask for Transformer based End-to-End Speech Recognition
• Data Movement Is All You Need: A Case Study on Optimizing Transformers
• Streaming End-to-end Speech Recognition For Mobile Devices
• Imputer: Sequence Modelling via Imputation and Dynamic Programming
• Improving Generalization of Transformer for Speech Recognition with Parallel Schedule Sampling and Relative Positional Embedding
• Low Latency End-to-End Streaming Speech Recognition with a Scout Network
• U2++: Unified Two-pass Bidirectional End-to-end Model for Speech Recognition
• Unsupervised Speech Recognition
• Sequence Discriminative Training for Deep Learning based Acoustic Keyword Spotting
• Radio2Text: Streaming Speech Recognition Using mmWave Radio Signals
• The Neural Noisy Channel
• Unsupervised pre-training for sequence to sequence speech recognition
• The ASRU 2019 Mandarin-English Code-Switching Speech Recognition Challenge: Open Datasets, Tracks, Methods and Results
• Gaussian Attention Model and Its Application to Knowledge Base Embedding and Question Answering
• Online Automatic Speech Recognition with Listen, Attend and Spell Model
• Towards a Competitive End-to-End Speech Recognition for CHiME-6 Dinner Party Transcription
• On Generalization Bounds of a Family of Recurrent Neural Networks
• A Simplified Fully Quantized Transformer for End-to-end Speech Recognition
• Aligned Contrastive Predictive Coding
• On the Comparison of Popular End-to-End Models for Large Scale Speech Recognition
• Why does CTC result in peaky behavior?
• Toward Fairness in Speech Recognition: Discovery and mitigation of performance disparities
• Deep Recurrent Convolutional Neural Network: Improving Performance For Speech Recognition
• Mixed Precision Low-bit Quantization of Neural Network Language Models for Speech Recognition
• Minimum Bayes Risk Training of RNN-Transducer for End-to-End Speech Recognition
• A Survey on Document-level Neural Machine Translation: Methods and Evaluation
• A Comparison of Label-Synchronous and Frame-Synchronous End-to-End Models for Speech Recognition
• WeNet: Production oriented Streaming and Non-streaming End-to-End Speech Recognition Toolkit
• ZeroPrompt: Streaming Acoustic Encoders are Zero-Shot Masked LMs
• Recurrent Neural Network Transducer for Audio-Visual Speech Recognition
• Completely Unsupervised Speech Recognition By A Generative Adversarial Network Harmonized With Iteratively Refined Hidden Markov Models
• Can Unconditional Language Models Recover Arbitrary Sentences?
• W2v-BERT: Combining Contrastive Learning and Masked Language Modeling for Self-Supervised Speech Pre-Training
• Plasma Confinement Mode Classification Using a Sequence-to-Sequence Neural Network With Attention
• PROCTER: PROnunciation-aware ConTextual adaptER for personalized speech recognition in neural transducers
• Streaming Chunk-Aware Multihead Attention for Online End-to-End Speech Recognition
• Efficient Training of Neural Transducer for Speech Recognition
• Improving Mandarin End-to-End Speech Recognition with Word N-gram Language Model
• Scaling End-to-End Models for Large-Scale Multilingual ASR
• Multi-head Monotonic Chunkwise Attention For Online Speech Recognition
• Learning Input and Recurrent Weight Matrices in Echo State Networks
• Machine learning classification of non-Markovian noise disturbing quantum dynamics
• Multi-level Contextual RNNs with Attention Model for Scene Labeling
• SANet: Structure-Aware Network for Visual Tracking
• A comparison of end-to-end models for long-form speech recognition
• Maximizing Mutual Information for Tacotron
• Dynamic Acoustic Unit Augmentation With BPE-Dropout for Low-Resource End-to-End Speech Recognition
• Emphasizing Unseen Words: New Vocabulary Acquisition for End-to-End Speech Recognition
• RNN-T Models Fail to Generalize to Out-of-Domain Audio: Causes and Solutions
• On-Device Personalization of Automatic Speech Recognition Models for Disordered Speech
• Listen with Intent: Improving Speech Recognition with Audio-to-Intent Front-End
• Improving performance of real-time full-band blind packet-loss concealment with predictive network
• Thank you for Attention: A survey on Attention-based Artificial Neural Networks for Automatic Speech Recognition
• Integrating Lattice-Free MMI into End-to-End Speech Recognition
• Magic dust for cross-lingual adaptation of monolingual wav2vec-2.0
• Efficient conformer-based speech recognition with linear attention
• Speech Recognition with Deep Recurrent Neural Networks
• Very Deep Convolutional Networks for End-to-End Speech Recognition
• Exploring Architectures, Data and Units For Streaming End-to-End Speech Recognition with RNN-Transducer
• Multi-Dialect Speech Recognition With A Single Sequence-To-Sequence Model
• Black Box Recursive Translations for Molecular Optimization
• CAT: CRF-based ASR Toolkit
• VoiceFilter-Lite: Streaming Targeted Voice Separation for On-Device Speech Recognition
• A Brief Introduction to Generative Models
• Universal ASR: Unifying Streaming and Non-Streaming ASR Using a Single Encoder-Decoder Model
• Automatic speech recognition for the Nepali language using CNN, bidirectional LSTM and ResNet
• Exploring Continuous Integrate-and-Fire for Adaptive Simultaneous Speech Translation
• Acoustic Data-Driven Subword Modeling for End-to-End Speech Recognition
• Incremental Text to Speech for Neural Sequence-to-Sequence Models using Reinforcement Learning
• Translating Math Formula Images to LaTeX Sequences Using Deep Neural Networks with Sequence-level Training
• From Senones to Chenones: Tied Context-Dependent Graphemes for Hybrid Speech Recognition
• Learning to Start for Sequence to Sequence Architecture
• Adversarial Attacks against Neural Networks in Audio Domain: Exploiting Principal Components
• Leveraging Phone Mask Training for Phonetic-Reduction-Robust E2E Uyghur Speech Recognition
• Deep context: end-to-end contextual speech recognition
• Developing Real-time Streaming Transformer Transducer for Speech Recognition on Large-scale Dataset
• Does Simultaneous Speech Translation need Simultaneous Models?
• Low-Latency Sequence-to-Sequence Speech Recognition and Translation by Partial Hypothesis Selection
• Unified End-to-End Speech Recognition and Endpointing for Fast and Efficient Speech Systems
• On the Impact of Word Error Rate on Acoustic-Linguistic Speech Emotion Recognition: An Update for the Deep Learning Era
• A Simple Baseline for Domain Adaptation in End to End ASR Systems Using Synthetic Data
• A Primal-Dual Method for Training Recurrent Neural Networks Constrained by the Echo-State Property
• SPGISpeech: 5,000 hours of transcribed financial audio for fully formatted end-to-end speech recognition
• Online Segment to Segment Neural Transduction
• UniSpeech at scale: An Empirical Study of Pre-training Method on Large-Scale Speech Recognition Dataset
• Streaming End-to-End Bilingual ASR Systems with Joint Language Identification
• Leveraging End-to-End Speech Recognition with Neural Architecture Search
• Fast End-to-End Speech Recognition via Non-Autoregressive Models and Cross-Modal Knowledge Transferring from BERT
• Federated Transfer Learning with Dynamic Gradient Aggregation
• Improved Conformer-based End-to-End Speech Recognition Using Neural Architecture Search
• Language model fusion for streaming end to end speech recognition
• Attention based end to end Speech Recognition for Voice Search in Hindi and English
• A study of latent monotonic attention variants
• Phoneme-Based Contextualization for Cross-Lingual Speech Recognition in End-to-End Models
• GapPredict: A Language Model for Resolving Gaps in Draft Genome Assemblies
• Gated Recurrent Context: Softmax-free Attention for Online Encoder-Decoder Speech Recognition
• Class LM and word mapping for contextual biasing in End-to-End ASR
• Transfer Learning Approaches for Streaming End-to-End Speech Recognition System
• Semi-supervised ASR by End-to-end Self-training
• Improving accuracy of rare words for RNN-Transducer through unigram shallow fusion
• End-to-End Speech Recognition: A review for the French Language
• Few-shot learning with attention-based sequence-to-sequence models
• On the quantization of recurrent neural networks
• Hessian-free Optimization for Learning Deep Multidimensional Recurrent Neural Networks
• How to Estimate Model Transferability of Pre-Trained Speech Models?
• Intermediate Loss Regularization for CTC-based Speech Recognition
• Attention as a Guide for Simultaneous Speech Translation
• The 2020 ESPnet update: new features, broadened applications, performance improvements, and future plans
• CIF: Continuous Integrate-and-Fire for End-to-End Speech Recognition
• Improved Mask-CTC for Non-Autoregressive End-to-End ASR
• Improving the Performance of Online Neural Transducer Models
• FastEmit: Low-latency Streaming ASR with Sequence-level Emission Regularization
• The Myths of Our Time: Fake News
• Neural Clustering Processes
• An Online Attention-based Model for Speech Recognition
• Exploring RNN-Transducer for Chinese Speech Recognition
• Two-pass Decoding and Cross-adaptation Based System Combination of End-to-end Conformer and Hybrid TDNN ASR Systems
• Streaming Object Detection for 3-D Point Clouds
• Attention-based Transducer for Online Speech Recognition
• Masked Pre-trained Encoder base on Joint CTC-Transformer
• A Unified Transferable Model for ML-Enhanced DBMS
• Morphological Inflection Generation with Hard Monotonic Attention
• An efficient and perceptually motivated auditory neural encoding and decoding algorithm for spiking neural networks
• Listen Attentively, and Spell Once: Whole Sentence Generation via a Non-Autoregressive Architecture for Low-Latency Speech Recognition
• Initial investigation of an encoder-decoder end-to-end TTS framework using marginalization of monotonic hard latent alignments
• A systematic comparison of grapheme-based vs. phoneme-based label units for encoder-decoder-attention models
• One In A Hundred: Select The Best Predicted Sequence from Numerous Candidates for Streaming Speech Recognition
• Synchronous Transformers for End-to-End Speech Recognition
• The Devil is in the Detail: Simple Tricks Improve Systematic Generalization of Transformers
• Multi-stage Progressive Compression of Conformer Transducer for On-device Speech Recognition
• Developing RNN-T Models Surpassing High-Performance Hybrid Models with Customization Capability
• Confidence-based Ensembles of End-to-End Speech Recognition Models
• Generating Natural Language Adversarial Examples through An Improved Beam Search Algorithm
• Evaluating Rewards for Question Generation Models
• Efficient Conformer with Prob-Sparse Attention Mechanism for End-to-EndSpeech Recognition
• Semantic Distance: A New Metric for ASR Performance Analysis Towards Spoken Language Understanding
• Improving RNN transducer with normalized jointer network
• Improving RNN Transducer Modeling for End-to-End Speech Recognition
• Private Language Model Adaptation for Speech Recognition
• Data Techniques For Online End-to-end Speech Recognition
• Exploring Machine Speech Chain for Domain Adaptation and Few-Shot Speaker Adaptation
• Research on Modeling Units of Transformer Transducer for Mandarin Speech Recognition
• Sequence Prediction with Neural Segmental Models
• Distilling the Knowledge of BERT for Sequence-to-Sequence ASR
• Adapting an Unadaptable ASR System
• RASR2: The RWTH ASR Toolkit for Generic Sequence-to-sequence Speech Recognition
• Streaming Small-Footprint Keyword Spotting using Sequence-to-Sequence Models
• Advancing CTC-CRF Based End-to-End Speech Recognition with Wordpieces and Conformers
• Bayesian Learning of LF-MMI Trained Time Delay Neural Networks for Speech Recognition
• Large-Scale Multilingual Speech Recognition with a Streaming End-to-End Model
• Equivalence of Segmental and Neural Transducer Modeling: A Proof of Concept
• Transformer with Bidirectional Decoder for Speech Recognition
• On the Inductive Bias of Word-Character-Level Multi-Task Learning for Speech Recognition
• An evaluation of word-level confidence estimation for end-to-end automatic speech recognition
• Injecting Text in Self-Supervised Speech Pretraining
• Real-time interactive sequence generation and control with Recurrent Neural Network ensembles
• Unsupervised Speaker Adaptation using Attention-based Speaker Memory for End-to-End ASR
• Towards Consistent Hybrid HMM Acoustic Modeling
• Exact Hard Monotonic Attention for Character-Level Transduction
• Speech Sentiment Analysis via Pre-trained Features from End-to-end ASR Models
• Multi-view Frequency LSTM: An Efficient Frontend for Automatic Speech Recognition
• Streaming Multi-speaker ASR with RNN-T
• Transformer-based Online Speech Recognition with Decoder-end Adaptive Computation Steps
• Semi-Supervised Learning with Data Augmentation for End-to-End ASR
• Neural Composition: Learning to Generate from Multiple Models
• End-to-end Anchored Speech Recognition
• Efficient Neural Architecture Search for End-to-end Speech Recognition via Straight-Through Gradients
• Sequence-level Speaker Change Detection with Difference-based Continuous Integrate-and-fire
• Powerful and Extensible WFST Framework for RNN-Transducer Losses
• Energy-Based Models with Applications to Speech and Language Processing
• SpeechColab Leaderboard: An Open-Source Platform for Automatic Speech Recognition Evaluation
• Learning to Discover, Ground and Use Words with Segmental Neural Language Models
• Reducing the gap between streaming and non-streaming Transducer-based ASR by adaptive two-stage knowledge distillation
• Advancing Acoustic-to-Word CTC Model with Attention and Mixed-Units
• PDAugment: Data Augmentation by Pitch and Duration Adjustments for Automatic Lyrics Transcription
• s-Transformer: Segment-Transformer for Robust Neural Speech Synthesis
• Reducing Streaming ASR Model Delay with Self Alignment
• Lost in Interpretation: Predicting Untranslated Terminology in Simultaneous Interpretation
• Multi-Stream End-to-End Speech Recognition
• A Streaming On-Device End-to-End Model Surpassing Server-Side Conventional Model Quality and Latency
• Contextualized Streaming End-to-End Speech Recognition with Trie-Based Deep Biasing and Shallow Fusion
• Self-Attention Aligner: A Latency-Control End-to-End Model for ASR Using Self-Attention Network and Chunk-Hopping
• Residual Energy-Based Models for End-to-End Speech Recognition
• Towards Fast and Accurate Streaming End-to-End ASR
• Confidence Estimation for Attention-based Sequence-to-sequence Models for Speech Recognition
• Do as I mean, not as I say: Sequence Loss Training for Spoken Language Understanding
• SimulLR: Simultaneous Lip Reading Transducer with Attention-Guided Adaptive Memory
• Tiny Transducer: A Highly-efficient Speech Recognition Model on Edge Devices
• Automatic Grammar Augmentation for Robust Voice Command Recognition
• Language model integration based on memory control for sequence to sequence speech recognition
• Speech Recognition by Simply Fine-tuning BERT
• A Better and Faster End-to-End Model for Streaming ASR
• Sequence-to-Sequence Modeling for Action Identification at High Temporal Resolution
• A spelling correction model for end-to-end speech recognition
• Tackling Sequence to Sequence Mapping Problems with Neural Networks
• A review of on-device fully neural end-to-end automatic speech recognition algorithms
• Unsupervised Speech Recognition via Segmental Empirical Output Distribution Matching
• An End-to-End Neural Network for Polyphonic Piano Music Transcription
• Dynamic Sparsity Neural Networks for Automatic Speech Recognition
• Dynamic latency speech recognition with asynchronous revision
• Minimum Bayes Risk Training for End-to-End Speaker-Attributed ASR
• End-to-End Code-Switching ASR for Low-Resourced Language Pairs
• Extending Recurrent Neural Aligner for Streaming End-to-End Speech Recognition in Mandarin
• Recurrent neural network transducer for Japanese and Chinese offline handwritten text recognition
• Faster, Simpler and More Accurate Hybrid ASR Systems Using Wordpieces
• Data-Driven Offline Optimization For Architecting Hardware Accelerators
• Oracle Teacher: Leveraging Target Information for Better Knowledge Distillation of CTC Models
• A Neural, Interactive-predictive System for Multimodal Sequence to Sequence Tasks
• End-to-End Automatic Speech Recognition Integrated With CTC-Based Voice Activity Detection
• End-to-end Whispered Speech Recognition with Frequency-weighted Approaches and Pseudo Whisper Pre-training
• Learning Diverse Stochastic Human-Action Generators by Learning Smooth Latent Transitions
• Listen, Attend, Spell and Adapt: Speaker Adapted Sequence-to-Sequence ASR
• Conv-Transformer Transducer: Low Latency, Low Frame Rate, Streamable End-to-End Speech Recognition
• Reducing Geographic Disparities in Automatic Speech Recognition via Elastic Weight Consolidation
• Bridging the gap between streaming and non-streaming ASR systems bydistilling ensembles of CTC and RNN-T models
• Streaming End-to-End ASR based on Blockwise Non-Autoregressive Models
• Using Synthetic Audio to Improve The Recognition of Out-Of-Vocabulary Words in End-To-End ASR Systems
• Transformer in action: a comparative study of transformer-based acoustic models for large scale speech recognition applications
• Improved Neural Language Model Fusion for Streaming Recurrent Neural Network Transducer
• Cross-attention conformer for context modeling in speech enhancement for ASR
• A Light-weight contextual spelling correction model for customizing transducer-based speech recognition systems
• Noisy Training Improves E2E ASR for the Edge
• The USTC-NELSLIP Systems for Simultaneous Speech Translation Task at IWSLT 2021
• On lattice-free boosted MMI training of HMM and CTC-based full-context ASR models
• Raw Waveform Encoder with Multi-Scale Globally Attentive Locally Recurrent Networks for End-to-End Speech Recognition
• Approximate Fixed-Points in Recurrent Neural Networks
• Adaptive Beam Search to Enhance On-device Abstractive Summarization
• Accelerator-Aware Training for Transducer-Based Speech Recognition
• Memory-augmented conformer for improved end-to-end long-form ASR
• Streaming Bilingual End-to-End ASR model using Attention over Multiple Softmax
• MixRep: Hidden Representation Mixup for Low-Resource Speech Recognition
• Internal Language Model Training for Domain-Adaptive End-to-End Speech Recognition
• Cross-utterance ASR Rescoring with Graph-based Label Propagation
• Guided contrastive self-supervised pre-training for automatic speech recognition
• Investigation of Training Label Error Impact on RNN-T
• Flexi-Transducer: Optimizing Latency, Accuracy and Compute forMulti-Domain On-Device Scenarios
• Improving Structured Text Recognition with Regular Expression Biasing
• Exploring Targeted Universal Adversarial Perturbations to End-to-end ASR Models
• Attention-based Neural Beamforming Layers for Multi-channel Speech Recognition
• On Addressing Practical Challenges for RNN-Transducer
• Head-synchronous Decoding for Transformer-based Streaming ASR
• High-dimensional sequence transduction
• Non-autoregressive Mandarin-English Code-switching Speech Recognition
• Exploring Pre-training with Alignments for RNN Transducer based End-to-End Speech Recognition
• CAT: A CTC-CRF based ASR Toolkit Bridging the Hybrid and the End-to-end Approaches towards Data Efficiency and Low Latency
• Incorporating VAD into ASR System by Multi-task Learning
• Minimum Latency Training Strategies for Streaming Sequence-to-Sequence ASR
• PyChain: A Fully Parallelized PyTorch Implementation of LF-MMI for End-to-End ASR
• Multitask Training with Text Data for End-to-End Speech Recognition
• Phoneme Based Neural Transducer for Large Vocabulary Speech Recognition
• On Minimum Word Error Rate Training of the Hybrid Autoregressive Transducer
• Improving Streaming Automatic Speech Recognition With Non-Streaming Model Distillation On Unsupervised Data
• Global-aware Beam Search for Neural Abstractive Summarization
• Efficient minimum word error rate training of RNN-Transducer for end-to-end speech recognition
• Simplified Self-Attention for Transformer-based End-to-End Speech Recognition
• Deliberation Model Based Two-Pass End-to-End Speech Recognition
• Parallel Rescoring with Transformer for Streaming On-Device Speech Recognition
• Characterization of Gravitational Waves Signals Using Neural Networks
• Integrating Source-channel and Attention-based Sequence-to-sequence Models for Speech Recognition
• Transformer ASR with Contextual Block Processing
• Domain Adaptation via Teacher-Student Learning for End-to-End Speech Recognition
• Modality Attention for End-to-End Audio-visual Speech Recognition
• End-To-End Speech Recognition Using A High Rank LSTM-CTC Based Model
• Statistical Parametric Speech Synthesis Using Bottleneck Representation From Sequence Auto-encoder
• Optimal Transport-based Alignment of Learned Character Representations for String Similarity
• Deep Shallow Fusion for RNN-T Personalization
• Improving RNN Transducer Based ASR with Auxiliary Tasks
• Dissecting User-Perceived Latency of On-Device E2E Speech Recognition
• Timers and Such: A Practical Benchmark for Spoken Language Understanding with Numbers
• FSR: Accelerating the Inference Process of Transducer-Based Models by Applying Fast-Skip Regularization
• Spectral Transform Forms Scalable Transformer
• A comparison of streaming models and data augmentation methods for robust speech recognition
• Speaker-aware speech-transformer
• Attention based on-device streaming speech recognition with large speech corpus
• Dynamic Encoder Transducer: A Flexible Solution For Trading Off Accuracy For Latency
• A Hybrid Recurrent Neural Network For Music Transcription
• Mutually-Constrained Monotonic Multihead Attention for Online ASR
• Developing neural machine translation models for Hungarian-English
• Streaming Speech-to-Confusion Network Speech Recognition
• Adversarially-Trained Normalized Noisy-Feature Auto-Encoder for Text Generation
• Learning When to Trust Which Teacher for Weakly Supervised ASR
• End-to-end Joint Punctuated and Normalized ASR with a Limited Amount of Punctuated Training Data
• Transformer Based Deliberation for Two-Pass Speech Recognition
• Improving RNN Transducer With Target Speaker Extraction and Neural Uncertainty Estimation
• Text-Only Domain Adaptation for End-to-End Speech Recognition through Down-Sampling Acoustic Representation
• Sequence-level self-learning with multiple hypotheses
• AV Taris: Online Audio-Visual Speech Recognition
• Character-Aware Attention-Based End-to-End Speech Recognition
• Speeding Up Neural Machine Translation Decoding by Cube Pruning
• Frame-level SpecAugment for Deep Convolutional Neural Networks in Hybrid ASR Systems
• Imitation Learning for Neural Morphological String Transduction
• Deep Learning Based Natural Language Processing for End to End Speech Translation
• CIF-based Collaborative Decoding for End-to-end Contextual Speech Recognition
• Minimum Word Error Rate Training with Language Model Fusion for End-to-End Speech Recognition
• Speaker Adaptation for End-to-End CTC Models
• Dynamic Gradient Aggregation for Federated Domain Adaptation
• A Click Sequence Model for Web Search
• Layer Pruning on Demand with Intermediate CTC
• Multi-mode Transformer Transducer with Stochastic Future Context
• CarneliNet: Neural Mixture Model for Automatic Speech Recognition
• Reducing Exposure Bias in Training Recurrent Neural Network Transducers
• Multi-Channel Transformer Transducer for Speech Recognition
• Sequence-to-Sequence Learning via Attention Transfer for Incremental Speech Recognition
• Cascade RNN-Transducer: Syllable Based Streaming On-device Mandarin Speech Recognition with a Syllable-to-Character Converter
• ChannelAugment: Improving generalization of multi-channel ASR by training with input channel randomization
• End-to-end Adaptation with Backpropagation through WFST for On-device Speech Recognition System
• Word-level confidence estimation for RNN transducers
• ASR Rescoring and Confidence Estimation with ELECTRA
• Integrating Categorical Features in End-to-End ASR
• Multitask Learning and Joint Optimization for Transformer-RNN-Transducer Speech Recognition
• Benchmarking LF-MMI, CTC and RNN-T Criteria for Streaming ASR
• Single-Queue Decoding for Neural Machine Translation
• A Unified Speaker Adaptation Approach for ASR
• Towards efficient end-to-end speech recognition with biologically-inspired neural networks
• Constructing Gradient Controllable Recurrent Neural Networks Using Hamiltonian Dynamics
• Integrating Knowledge into End-to-End Speech Recognition from External Text-Only Data
• Context-Aware Transformer Transducer for Speech Recognition
• A practical two-stage training strategy for multi-stream end-to-end speech recognition
• End-to-End Code Switching Language Models for Automatic Speech Recognition
• Recurrent Neural Network for End-to-End Modeling of Laminar-Turbulent Transition
• HMM-Free Encoder Pre-Training for Streaming RNN Transducer
• Extremely Low Footprint End-to-End ASR System for Smart Device