Language Model Data Augmentation for Keyword Spotting in Low-Resourced Training Conditions

Gorin, Arseniy; Lileikytė, Rasa; Huang, Guangpu; Lamel, Lori; Gauvain, Jean‐Luc; Laurent, Antoine

doi:10.21437/interspeech.2016-1200

Cited by 11 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The FCN and CNSA show similar trends across all experimental conditions, therefore we can perhaps infer that both of these classifiers are overfitting the training data. Since they are overfitting, we would normally expect to see improvements on these classifiers by applying web data augmentation ( Lamel et al, 2002 ; Gorin et al, 2016 ). However, we barely gained any benefit from data augmentation, apparently because the in-domain and out-domain datasets had different acoustic feature distributions.…”

Section: Discussionmentioning

confidence: 99%

Analysis of acoustic and voice quality features for the classification of infant and mother vocalizations

Hasegawa‐Johnson

McElwain

2021

Speech Communication

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Analysis of acoustic and voice quality features for the classification of infant and mother vocalizations

Hasegawa‐Johnson

McElwain

2021

Speech Communication

View full text Add to dashboard Cite

“…Data augmentation provides a potential solution for data scarcity because it is less time-consuming than collecting and transcribing real speech data and people have shown in many contexts [4,5,6,7,8] that it improves results. Text generation -a data augmentation method -has been proposed in [9,10,11,12,13,14,15] with the aim of improving language models and therefore automatic speech recognition performance for CS speech.…”

Section: Introductionmentioning

confidence: 99%

Improving Code-Switching Language Modeling with Artificially Generated Texts Using Cycle-Consistent Adversarial Networks

Li¹,

Vu²

2020

Interspeech 2020

View full text Add to dashboard Cite

This paper presents our latest effort on improving Codeswitching language models that suffer from data scarcity. We investigate methods to augment Code-switching training text data by artificially generating them. Concretely, we propose a cycle-consistent adversarial networks based framework to transfer monolingual text into Code-switching text, considering Code-switching as a speaking style. Our experimental results on the SEAME corpus show that utilizing artificially generated Code-switching text data improves consistently the language model as well as the automatic speech recognition performance.

show abstract

“…Data augmentation via SMT has been explored in the past for keyword spotting [1] and ASR [8,9]. These studies primarily focus on incorporating raw translation output as a component in the LM.…”

Section: Introductionmentioning

confidence: 99%

“…There has been a growing interest in the area of data augmentation for ASR language modeling. Previous studies include training a recurrent neural network (RNN) based LM on transcriptions and using it to generate synthetic samples for augmentation [1]. SeqGAN, a generative adversarial model for sequences, has been employed for pretraining a code-switched LM [2].…”

Section: Introductionmentioning

confidence: 99%

Language Model Bootstrapping Using Neural Machine Translation for Conversational Speech Recognition

Punjabi¹,

Arsikere²,

Garimella³

2019

2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)

View full text Add to dashboard Cite

Building conversational speech recognition systems for new languages is constrained by the availability of utterances capturing user-device interactions. Data collection is expensive and limited by speed of manual transcription. In order to address this, we advocate the use of neural machine translation as a data augmentation technique for bootstrapping language models. Machine translation (MT) offers a systematic way of incorporating collections from mature, resource-rich conversational systems that may be available for a different language. However, ingesting raw translations from a general purpose MT system may not be effective owing to the presence of named entities, intra sentential code-switching and the domain mismatch between the conversational data being translated and the parallel text used for MT training. To circumvent this, we explore following domain adaptation techniques: (a) sentence embedding based data selection for MT training, (b) model finetuning, and (c) rescoring and filtering translated hypotheses. Using Hindi language as the experimental testbed, we supplement transcribed collections with translated US English utterances. We observe a relative word error rate reduction of 7.8-15.6%, depending on the bootstrapping phase. Fine grained analysis reveals that translation particularly aids the interaction scenarios underrepresented in the transcribed data.

show abstract

Language Model Data Augmentation for Keyword Spotting in Low-Resourced Training Conditions

Cited by 11 publications

References 21 publications

Analysis of acoustic and voice quality features for the classification of infant and mother vocalizations

Analysis of acoustic and voice quality features for the classification of infant and mother vocalizations

Improving Code-Switching Language Modeling with Artificially Generated Texts Using Cycle-Consistent Adversarial Networks

Language Model Bootstrapping Using Neural Machine Translation for Conversational Speech Recognition

Contact Info

Product

Resources

About