Proceedings ArticleDOI
Purely Sequence-Trained Neural Networks for ASR Based on Lattice-Free MMI.
Daniel Povey,Vijayaditya Peddinti,Daniel Galvez,Pegah Ghahremani,Vimal Manohar,Xingyu Na,Yiming Wang,Sanjeev Khudanpur +7 more
- pp 2751-2755
Reads0
Chats0
TLDR
A method to perform sequencediscriminative training of neural network acoustic models without the need for frame-level cross-entropy pre-training is described, using the lattice-free version of the maximum mutual information (MMI) criterion: LF-MMI.Abstract:
In this paper we describe a method to perform sequencediscriminative training of neural network acoustic models without the need for frame-level cross-entropy pre-training. We use the lattice-free version of the maximum mutual information (MMI) criterion: LF-MMI. To make its computation feasible we use a phone n-gram language model, in place of the word language model. To further reduce its space and time complexity we compute the objective function using neural network outputs at one third the standard frame rate. These changes enable us to perform the computation for the forward-backward algorithm on GPUs. Further the reduced output frame-rate also provides a significant speed-up during decoding. We present results on 5 different LVCSR tasks with training data ranging from 100 to 2100 hours. Models trained with LFMMI provide a relative word error rate reduction of ∼11.5%, over those trained with cross-entropy objective function, and ∼8%, over those trained with cross-entropy and sMBR objective functions. A further reduction of ∼2.5%, relative, can be obtained by fine tuning these models with the word-lattice based sMBR objective function.read more
Citations
More filters
Proceedings ArticleDOI
SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
TL;DR: This work presents SpecAugment, a simple data augmentation method for speech recognition that is applied directly to the feature inputs of a neural network (i.e., filter bank coefficients) and achieves state-of-the-art performance on the LibriSpeech 960h and Swichboard 300h tasks, outperforming all prior work.
Proceedings ArticleDOI
ESPNet: End-to-end speech processing toolkit
Shinji Watanabe,Takaaki Hori,Shigeki Karita,Tomoki Hayashi,Jiro Nishitoba,Yuya Unno,Nelson Yalta,Jahn Heymann,Matthew Wiesner,Nanxin Chen,Adithya Renduchintala,Tsubasa Ochiai +11 more
TL;DR: In this article, a new open source platform for end-to-end speech processing named ESPnet is introduced, which mainly focuses on automatic speech recognition (ASR), and adopts widely used dynamic neural network toolkits, Chainer and PyTorch, as a main deep learning engine.
Proceedings ArticleDOI
A study on data augmentation of reverberant speech for robust speech recognition
TL;DR: It is found that the performance gap between using simulated and real RIRs can be eliminated when point-source noises are added, and the trained acoustic models not only perform well in the distant- talking scenario but also provide better results in the close-talking scenario.
Journal ArticleDOI
Hybrid CTC/Attention Architecture for End-to-End Speech Recognition
TL;DR: The proposed hybrid CTC/attention end-to-end ASR is applied to two large-scale ASR benchmarks, and exhibits performance that is comparable to conventional DNN/HMM ASR systems based on the advantages of both multiobjective learning and joint decoding without linguistic resources.
Posted Content
Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfaces
Alice Coucke,Alaa Saade,Adrien Ball,Theodore Bluche,Alexandre Caulier,David Leroy,Clément Doumouro,Thibault Gisselbrecht,Francesco Caltagirone,Thibaut Lavril,Maël Primet,Joseph Dureau +11 more
TL;DR: The machine learning architecture of the Snips Voice Platform is presented, a software solution to perform Spoken Language Understanding on microprocessors typical of IoT devices that is fast and accurate while enforcing privacy by design, as no personal user data is ever collected.
References
More filters
Proceedings ArticleDOI
Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks
TL;DR: This paper presents a novel method for training RNNs to label unsegmented sequences directly, thereby solving both problems of sequence learning and post-processing.
Posted Content
Deep Speech: Scaling up end-to-end speech recognition
Awni Hannun,Carl Case,Jared Casper,Bryan Catanzaro,Greg Diamos,Erich Elsen,Ryan Prenger,Sanjeev Satheesh,Shubho Sengupta,Adam Coates,Andrew Y. Ng +10 more
TL;DR: Deep Speech, a state-of-the-art speech recognition system developed using end-to-end deep learning, outperforms previously published results on the widely studied Switchboard Hub5'00, achieving 16.0% error on the full test set.
Proceedings ArticleDOI
Audio augmentation for speech recognition.
TL;DR: This paper investigates audio-level speech augmentation methods which directly process the raw signal, and presents results on 4 different LVCSR tasks with training data ranging from 100 hours to 1000 hours, to examine the effectiveness of audio augmentation in a variety of data scenarios.
Proceedings ArticleDOI
A time delay neural network architecture for efficient modeling of long temporal contexts.
TL;DR: This paper proposes a time delay neural network architecture which models long term temporal dependencies with training times comparable to standard feed-forward DNNs and uses sub-sampling to reduce computation during training.
Book ChapterDOI
An n log n algorithm for minimizing states in a finite automaton
TL;DR: An algorithm is given for minimizing the number of states in a finite automaton or for determining if two finite automata are equivalent and the running time is bounded by k n log n.