Selective Deep Speaker Embedding Enhancement for Speaker Verification

Jung, Jee-weon; Kim, Ju-Ho; Shim, Hye-jin; Kim, Seung-bin; Yu, Ha-Jin

doi:10.21437/odyssey.2020-25

Cited by 4 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [17], an i-vector [3] denoising method based on statistical models was proposed and, subsequently, extended to x-vectors [1] using a denoising autoencoder [18]. Concurrently, DNNs were employed to further disentangle speaker-relevant andirrelevant factors from speaker embeddings [19]- [21]. In [22], a DNN was used to learn input-specific magnitude scalers for speaker embeddings, which, in turn, yielded nonmonotonically calibrated cosine similarity scores with improved discrimination ability.…”

Section: Introductionmentioning

confidence: 99%

Task-specific Optimization of Virtual Channel Linear Prediction-based Speech Dereverberation Front-End for Far-Field Speaker Verification

Yang¹,

Chang²

2021

Preprint

View full text Add to dashboard Cite

Developing a single-microphone speech denoising or dereverberation front-end for robust automatic speaker verification (ASV) in noisy far-field speaking scenarios is challenging. To address this problem, we present a novel front-end design that involves a recently proposed extension of the weighted prediction error (WPE) speech dereverberation algorithm, the virtual acoustic channel expansion (VACE)-WPE. It is demonstrated experimentally in this study that unlike the conventional WPE algorithm, the VACE-WPE can be explicitly trained to cancel out both late reverberation and background noise. To build the front-end, the VACE-WPE is first independently (pre)trained to produce "noisy" dereverberated signals. Subsequently, given a pretrained speaker embedding model, the VACE-WPE is additionally fine-tuned within a task-specific optimization (TSO) framework, causing the speaker embedding extracted from the processed signal to be similar to that extracted from the "noisefree" target signal. Moreover, to extend the application of the proposed front-end to more general, unconstrained "in-the-wild" ASV scenarios beyond controlled far-field conditions, we propose a distortion regularization method for the VACE-WPE within the TSO framework. The effectiveness of the proposed approach is verified on both far-field and in-the-wild ASV benchmarks, demonstrating its superiority over fully neural front-ends and other TSO methods in various cases.

show abstract