The voice bank corpus: Design, collection and data analysis of a large regional accent speech database

By Christophe Veaux, Junichi Yamagishi, Simon King. 2013

478 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Experimental and Cognitive Psychology, Physiology, Evaluation & Benchmarks, Speech & Audio

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Experimental and Cognitive Psychology

In Physiology

In Evaluation & Benchmarks

In Speech & Audio

Citations in the corpus

Audio Codec Enhancement with Generative Adversarial Networks (2020, methodology): “The dataset we used for speech [25] consisted of 30 speakers (15 male, 15 female) from the clean Voice Bank [26] dataset (popularly known as the VCTK).”
A Wavenet for Speech Denoising (2017, methodology): “The dataset we used [12, 18] was generated from two sources: speech data was supplied by the Voice Bank corpus [19] while environmental sounds were provided by the Diverse Environments Multichannel Acoustic Noise Database (DEMAND) [20].”
SEGAN: Speech Enhancement Generative Adversarial Network (2017, neutral): “4), becomes”
Neural Analysis and Synthesis: Reconstructing Speech from Self-Supervised Representations (2021, methodology): “Specifically, we averaged the representation of each layer along the time-axis and visualized utterances of 20 randomly selected speakers from the VCTK dataset using TSNE [46, 45].”
Sergan: Speech Enhancement Using Relativistic Generative Adversarial Networks with Gradient Penalty (2019, methodology): “The database is derived from the voice bank corpus [23] from which recordings from 28 speakers were chosen for the training set ( 11 572 utterances) and 2 for the test set ( 824 utterances).”
Investigating RNN-based speech enhancement methods for noise-robust Text-to-Speech (2016, unknown): “We selected from the Voice Bank corpus [24] 28 speakers - 14 male and 14 female of the same accent region (England) and another 56 speakers - 28 male and 28 female - of different accent regions (Scotland and United States).”
SpeechBrain: A General-Purpose Speech Toolkit (2021, methodology): “• Voicebank-DEMAND [96]: It contains speech of 30 clean speakers extracted from the Voice Bank corpus [41]: 28 are included in the training set, and two are in the validation set.”
Phase-aware Speech Enhancement with Deep Complex U-Net (2019, methodology): “Noise and clean speech recordings were provided from the Diverse Environments Multichannel Acoustic Noise Database (DEMAND) (Thiemann et al., 2013) and the Voice Bank corpus (Veaux et al., 2013), respectively, each recorded with sampling rate of 48kHz.”
Deep Residual-Dense Lattice Network for Speech Enhancement (2020, methodology): “The test set included 824 clean speech recordings of two speakers from the Voice Bank corpus (393 from 𝑝 232 and 431 from 𝑝 257 ) (Veaux, Yamagishi, and King 2013).”
Separate Anything You Describe (2023, unknown): “The Voicebank-DEMAND data-set [67] integrates the Voicebank dataset [67], which includes clean speech, and the DEMAND [68] dataset, which encompasses a variety of background sounds that are used to create noisy speech.”
Speech Enhancement for a Noise-Robust Text-to-Speech Synthesis System Using Deep Recurrent Neural Networks (2016, methodology): “We selected from the Voice Bank corpus [23] 28 speakers - 14 male and 14 female of the same accent region (England) and another 56 speakers - 28 male and 28 female - of different accent regions (Scotland and United States).”
Improving GANs for Speech Enhancement (2020, methodology): “The database is originated from the Voice Bank corpus [29] and consists of data from 30 speakers.”
Speech Enhancement for Noise-Robust Speech Synthesis Using Wasserstein GAN (2019, methodology): “To train WGAN, we used 30 speakers (15 male and 15 female speakers) from voice bank corpus [18].”
Exploring the Best Loss Function for DNN-Based Low-latency Speech Enhancement with Temporal Convolutional Networks (2020, methodology): “We conduct experiments on two different datasets: the VoiceBank + DEMAND (VBD) dataset1 [19–21], which is relatively small, and the DNS Challenge dataset [1], which is a large-scale”
SUPERB-SG: Enhanced Speech processing Universal PERformance Benchmark for Semantic and Generative Capabilities (2022, unknown): “In SUPERB-SG, we discuss the speech enhancement problem on the VoicebankDEMAND (Veaux et al., 2013) (CC BY 4.”
Time-Frequency Masking-Based Speech Enhancement Using Generative Adversarial Network (2018, unknown): “The dataset comprises of 30 speakers from the Voice Bank corpus [21].”
A study of speaker adaptation for DNN-based speech synthesis (2015, methodology): “In the experiments, we used the Voice Bank corpus [28] to assess the performance of the adaptation techniques.”
NADiffuSE: Noise-aware Diffusion-based Model for Speech Enhancement (2023, methodology): “We simulated the new dataset using VoiceBank [40] and Noisex-92 with the same setting in Section IV-A1.”