Search Results for author:
Found 48 papers, 5 papers with code
Self-Supervised Speech Representation Learning: A Review
Although self-supervised speech representation is still a nascent research area, it is closely related to acoustic word embedding and learning with zero lexical resources, both of which have seen active research for many years.
E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR
Improving the performance of end-to-end ASR models on long utterances ranging from minutes to hours in length is an ongoing challenge in speech recognition.
Improving Rare Word Recognition with LM-aware MWER Training
Language models (LMs) significantly improve the recognition accuracy of end-to-end (E2E) models on words rarely seen during training, when used in either the shallow fusion or the rescoring setups.
Streaming Align-Refine for Non-autoregressive Deliberation
We propose a streaming non-autoregressive (non-AR) decoding algorithm to deliberate the hypothesis alignment of a streaming RNN-T model.
A Unified Cascaded Encoder ASR Model for Dynamic Model Sizes
In this paper, we propose a dynamic cascaded encoder Automatic Speech Recognition (ASR) model, which unifies models for different deployment scenarios.
Sentence-Select: Large-Scale Language Model Data Selection for Rare-Word Speech Recognition
We down-select a large corpus of web search queries by a factor of 53x and achieve better LM perplexities than without down-selection.
Improving the fusion of acoustic and text representations in RNN-T
The recurrent neural network transducer (RNN-T) has recently become the mainstream end-to-end approach for streaming automatic speech recognition (ASR).
Joint Unsupervised and Supervised Training for Multilingual ASR
Our average WER of all languages outperforms average monolingual baseline by 33. 3%, and the state-of-the-art 2-stage XLSR by 32%.
BigSSL: Exploring the Frontier of Large-Scale Semi-Supervised Learning for Automatic Speech Recognition
We summarize the results of a host of efforts using giant automatic speech recognition (ASR) models pre-trained using large, diverse unlabeled datasets containing approximately a million hours of audio.
Tied & Reduced RNN-T Decoder
Previous works on the Recurrent Neural Network-Transducer (RNN-T) models have shown that, under some conditions, it is possible to simplify its prediction network with little or no loss in recognition accuracy (arXiv:2003. 07705 [eess. AS], [2], arXiv:2012. 06749 [cs. CL]).
Scaling End-to-End Models for Large-Scale Multilingual ASR
Building ASR models across many languages is a challenging multi-task learning problem due to large variations and heavily unbalanced data.
Lookup-Table Recurrent Language Models for Long Tail Speech Recognition
We introduce Lookup-Table Language Models (LookupLM), a method for scaling up the size of RNN language models with only a constant increase in the floating point operations, by increasing the expressivity of the embedding table.
Learning Word-Level Confidence For Subword End-to-End ASR
We study the problem of word-level confidence estimation in subword-based end-to-end (E2E) models for automatic speech recognition (ASR).
Transformer Based Deliberation for Two-Pass Speech Recognition
In this work, we explore using transformer layers instead of long-short term memory (LSTM) layers for deliberation rescoring.
Less Is More: Improved RNN-T Decoding Using Limited Label Context and Path Merging
End-to-end models that condition the output label sequence on all previously predicted labels have emerged as popular alternatives to conventional systems for automatic speech recognition (ASR).
A Better and Faster End-to-End Model for Streaming ASR
To address this, we explore replacing the LSTM layers in the encoder of our E2E model with Conformer layers [4], which has shown good improvements for ASR.
Audio and Speech Processing Sound
Cascaded encoders for unifying streaming and non-streaming ASR
The proposed model consists of streaming and non-streaming encoders.
Multitask Training with Text Data for End-to-End Speech Recognition
We propose a multitask training method for attention-based end-to-end speech recognition models.
FastEmit: Low-latency Streaming ASR with Sequence-level Emission Regularization
FastEmit also improves streaming ASR accuracy from 4. 4%/8. 9% to 3. 1%/7. 5% WER, meanwhile reduces 90th percentile latency from 210 ms to only 30 ms on LibriSpeech.
Dual-mode ASR: Unify and Improve Streaming ASR with Full-context Modeling
Streaming automatic speech recognition (ASR) aims to emit each hypothesized word as quickly and accurately as possible, while full-context ASR waits for the completion of a full speech utterance before emitting completed hypotheses.
Improving Tail Performance of a Deliberation E2E ASR Model Using a Large Text Corpus
End-to-end (E2E) automatic speech recognition (ASR) systems lack the distinct language model (LM) component that characterizes traditional speech systems.
Improving Proper Noun Recognition in End-to-End ASR By Customization of the MWER Loss Criterion
Proper nouns present a challenge for end-to-end (E2E) automatic speech recognition (ASR) systems in that a particular name may appear only rarely during training, and may have a pronunciation similar to that of a more common word.
RNN-T Models Fail to Generalize to Out-of-Domain Audio: Causes and Solutions
On a long-form YouTube test set, when the nonstreaming RNN-T model is trained with shorter segments of data, the proposed combination improves word error rate (WER) from 22. 3% to 14. 8%; when the streaming RNN-T model trained on short Search queries, the proposed techniques improve WER on the YouTube set from 67. 0% to 25. 3%.
Towards Fast and Accurate Streaming End-to-End ASR
RNN-T EP+LAS, together with MWER training brings in 18. 7% relative WER reduction and 160ms 90-percentile latency reductions compared to the original proposed RNN-T EP model.
Audio and Speech Processing
A Streaming On-Device End-to-End Model Surpassing Server-Side Conventional Model Quality and Latency
Thus far, end-to-end (E2E) models have not been shown to outperform state-of-the-art conventional models with respect to both quality, i. e., word error rate (WER), and latency, i. e., the time the hypothesis is finalized after the user stops speaking.
Deliberation Model Based Two-Pass End-to-End Speech Recognition
End-to-end (E2E) models have made rapid progress in automatic speech recognition (ASR) and perform competitively relative to conventional models.
Recognizing long-form speech using streaming end-to-end models
In this work, we examine the ability of E2E models to generalize to unseen domains, where we find that models trained on short utterances fail to generalize to long-form speech.
Large-Scale Multilingual Speech Recognition with a Streaming End-to-End Model
Multilingual end-to-end (E2E) models have shown great promise in expansion of automatic speech recognition (ASR) coverage of the world's languages.
Two-Pass End-to-End Speech Recognition
However, this model still lags behind a large state-of-the-art conventional model in quality [2].
Improving Performance of End-to-End ASR on Numeric Sequences
This out-of-vocabulary (OOV) issue is addressed in conventional ASR systems by training part of the model on spoken domain utterances (e. g.
Phoneme-Based Contextualization for Cross-Lingual Speech Recognition in End-to-End Models
Contextual automatic speech recognition, i. e., biasing recognition towards a given context (e. g. user's playlists, or contacts), is challenging in end-to-end (E2E) models.
A spelling correction model for end-to-end speech recognition
Attention-based sequence-to-sequence models for speech recognition jointly train an acoustic model, language model (LM), and alignment mechanism using a single neural network and require only parallel audio-text pairs.
Streaming End-to-end Speech Recognition For Mobile Devices
End-to-end (E2E) models, which directly predict output character sequences given input speech, are good candidates for on-device speech recognition.
Contextual Speech Recognition with Difficult Negative Training Examples
Improving the representation of contextual information is key to unlocking the potential of end-to-end (E2E) automatic speech recognition (ASR).
Deep context: end-to-end contextual speech recognition
Our approach, which we re- fer to as Contextual Listen, Attend and Spell (CLAS) jointly- optimizes the ASR components along with embeddings of the context n-grams.
A Comparison of Techniques for Language Model Integration in Encoder-Decoder Speech Recognition
In this paper, we compare a suite of past methods and some of our own proposed methods for using unpaired text data to improve encoder-decoder models.
An analysis of incorporating an external language model into a sequence-to-sequence model
Attention-based sequence-to-sequence models for automatic speech recognition jointly train an acoustic model, language model, and alignment mechanism.
State-of-the-art Speech Recognition With Sequence-to-Sequence Models
Attention-based encoder-decoder architectures such as Listen, Attend, and Spell (LAS), subsume the acoustic, pronunciation and language model components of a traditional automatic speech recognition (ASR) system into a single neural network.
Improving the Performance of Online Neural Transducer Models
Neural transducer is a streaming sequence-to-sequence model, but has shown a significant degradation in performance compared to non-streaming models such as Listen, Attend and Spell (LAS).
Multi-Dialect Speech Recognition With A Single Sequence-To-Sequence Model
Sequence-to-sequence models provide a simple and elegant solution for building speech recognition systems by folding separate components of a typical system, namely acoustic (AM), pronunciation (PM) and language (LM) models into a single neural network.
Minimum Word Error Rate Training for Attention-based Sequence-to-Sequence Models
Sequence-to-sequence models, such as attention-based models in automatic speech recognition (ASR), are typically trained to optimize the cross-entropy criterion which corresponds to improving the log-likelihood of the data.
No Need for a Lexicon? Evaluating the Value of the Pronunciation Lexica in End-to-End Models
However, there has been little previous work comparing phoneme-based versus grapheme-based sub-word units in the end-to-end modeling framework, to determine whether the gains from such approaches are primarily due to the new probabilistic model, or from the joint learning of the various components with grapheme-based units.
Multilingual Speech Recognition With A Single End-To-End Model
Training a conventional automatic speech recognition (ASR) system to support multiple languages is challenging because the sub-word unit, lexicon and word inventories are typically language specific.
Learning Compact Recurrent Neural Networks
Recurrent neural networks (RNNs), including long short-term memory (LSTM) RNNs, have produced state-of-the-art results on a variety of speech recognition tasks.
Structured Transforms for Small-Footprint Deep Learning
We consider the task of building compact deep learning pipelines suitable for deployment on storage and power constrained mobile devices.
Improvements to deep convolutional neural networks for LVCSR
We find that with these improvements, particularly with fMLLR and dropout, we are able to achieve an additional 2-3% relative improvement in WER on a 50-hour Broadcast News task over our previous best CNN baseline.
Accelerating Hessian-free optimization for deep neural networks by implicit preconditioning and sampling
This study aims at speeding up Hessian-free training, both by means of decreasing the amount of data used for training, as well as through reduction of the number of Krylov subspace solver iterations used for implicit estimation of the Hessian.
