Improved performance of Aurora 4 using HTK and unsupervised MLLR adaptation

Yeung, Siu-Kei Au; Siu, Man-Hung

doi:10.21437/interspeech.2004-108

Cited by 26 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [6] DAT is employed for noise adaptation on a noise corrupted version of WSJ [56] as the target dataset. Using the Aurora-4 [57] dataset which has labels associated to the noise type, Serdyuk et al [33] train an adversarial noise classifier. In [8] and [39] DAT is utilized for accent adaptation for Mandarin and English respectively.…”

Section: Domain Adversarial Trainingmentioning

confidence: 99%

Sample-Efficient Unsupervised Domain Adaptation of Speech Recognition Systems: A case study for Modern Greek

Paraskevopoulos¹,

Kouzelis²,

Rouvalis³

et al. 2023

Preprint

View full text Add to dashboard Cite

<p>Modern speech recognition systems exhibits rapid performance degradation under domain shift. This issue is especially prevalent in data-scarce settings, such as low-resource languages, where diversity of training data is limited.</p> <p>In this work we propose M2DS2, a simple and sample-efficient finetuning strategy for large pretrained speech models, based on mixed source and target domain self-supervision. We find that including source domain self-supervision stabilizes training and avoids mode collapse of the latent representations. For evaluation, we collect HParl, a 120 hour speech corpus for Greek, consisting of plenary sessions in the Greek Parliament. We merge HParl with two popular Greek corpora to create GREC-MD, a test-bed for multi-domain evaluation of Greek ASR systems. In our experiments we find that, while other Unsupervised Domain Adaptation baselines fail in this resource-constrained environment, M2DS2 yields significant improvements for cross-domain adaptation, even when a only a few hours of in-domain audio are available. When we relax the problem in a weakly supervised setting, we find that independent adaptation for audio using M2DS2 and language using simple LM augmentation techniques is particularly effective, yielding word error rates comparable to the fully supervised baselines.</p>

show abstract

Section: Domain Adversarial Trainingmentioning

confidence: 99%

Sample-Efficient Unsupervised Domain Adaptation of Speech Recognition Systems: A case study for Modern Greek

Paraskevopoulos¹,

Kouzelis²,

Rouvalis³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…We used the SDM recordings for our experiments, which is ideal for single-channel far-field evaluation. Aurora-4 [24] is an 80hour close-talk speech corpus based on Wall Street Journal (WSJ) data with artificially added noise. We evaluated on the eval92 and 0166 subsets.…”

Section: Datasets and Modelmentioning

confidence: 99%

Frustratingly Easy Noise-aware Training of Acoustic Models

Raj,

Villalba,

Povey

et al. 2020

Preprint

View full text Add to dashboard Cite

Environmental noises and reverberation have a detrimental effect on the performance of automatic speech recognition (ASR) systems. Multi-condition training of neural networkbased acoustic models is used to deal with this problem, but it requires many-folds data augmentation, resulting in increased training time. In this paper, we propose utterance-level noise vectors for noise-aware training of acoustic models in hybrid ASR. Our noise vectors are obtained by combining the means of speech frames and silence frames in the utterance, where the speech/silence labels may be obtained from a GMM-HMM model trained for ASR alignments, such that no extra computation is required beyond averaging of feature vectors. We show through experiments on AMI and Aurora-4 that this simple adaptation technique can result in 6-7% relative WER improvement. We implement several embedding-based adaptation baselines proposed in literature, and show that our method outperforms them on both the datasets. Finally, we extend our method to the online ASR setting by using frame-level maximum likelihood for the mean estimation.

show abstract