Speaker Diarization Using Deep Recurrent Convolutional Neural Networks for Speaker Embeddings

Cyrta, Paweł; Trzciński, Tomasz Piotr; Stokowiec, Wojciech

doi:10.1007/978-3-319-67220-5_10

Cited by 23 publications

(22 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…AMI: To evaluate the performance of the proposed embeddings on the speaker diarization task, we use a subset of the AMI meeting corpus [19] that is frequently used for evaluating diarization performance [20,21]. It consists of audio recordings from 26 meetings.…”

Section: Datasetsmentioning

confidence: 99%

Robust Speaker Recognition Using Unsupervised Adversarial Invariance

Peri

Pal

Jati

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

In this paper, we address the problem of speaker recognition in challenging acoustic conditions using a novel method to extract robust speaker-discriminative speech representations. We adopt a recently proposed unsupervised adversarial invariance architecture to train a network that maps speaker embeddings extracted using a pretrained model onto two lower dimensional embedding spaces. The embedding spaces are learnt to disentangle speaker-discriminative information from all other information present in the audio recordings, without supervision about the acoustic conditions. We analyze the robustness of the proposed embeddings to various sources of variability present in the signal for speaker verification and unsupervised clustering tasks on a large-scale speaker recognition corpus. Our analyses show that the proposed system substantially outperforms the baseline in a variety of challenging acoustic scenarios. Furthermore, for the task of speaker diarization on a real-world meeting corpus, our system shows a relative improvement of 36% in the diarization error rate compared to the state-of-the-art baseline.

show abstract

Section: Datasetsmentioning

confidence: 99%

Robust Speaker Recognition Using Unsupervised Adversarial Invariance

Peri

Pal

Jati

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…While we do not study the mediation of meaning, we do claim to capture "those elements of speech that are not embeddings has been approached with different neural network architectures such as Siamese [6], fully connected [43], and CNN [24,25,10]. Only very recently have RNNs been used successfully [33,23,44].…”

Section: Related Workmentioning

confidence: 99%

“…Cyrta et al [33] suggest to learn the embeddings by training a recurrent convolutional neural network for the task of speaker classification. Although they utilize recurrent layers to retrieve temporal information, the feature extraction is still done by convolutional layers.…”

Section: Related Workmentioning

confidence: 99%

Capturing Suprasegmental Features of a Voice with RNNs for Improved Speaker Clustering

Stadelmann

Glinski-Haefeli

Gerber

et al. 2018

Artificial Neural Networks in Pattern Recognition

View full text Add to dashboard Cite

Abstract. Deep neural networks have become a veritable alternative to classic speaker recognition and clustering methods in recent years. However, while the speech signal clearly is a time series, and despite the body of literature on the benefits of prosodic (suprasegmental) features, identifying voices has usually not been approached with sequence learning methods. Only recently has a recurrent neural network (RNN) been successfully applied to this task, while the use of convolutional neural networks (CNNs) (that are not able to capture arbitrary time dependencies, unlike RNNs) still prevails. In this paper, we show the effectiveness of RNNs for speaker recognition by improving state of the art speaker clustering performance and robustness on the classic TIMIT benchmark. We provide arguments why RNNs are superior by experimentally showing a "sweet spot" of the segment length for successfully capturing prosodic information that has been theoretically predicted in previous work.

show abstract

“…In [16], a conditional variational autoencoder is proposed in order to generate prosodic features for speech synthesis by sampling prosodic embeddings from the bottleneck representation. Specifically for separation tasks, speaker-discriminative embeddings are produced for targeted voice separation in [6] and for diarization in [17] yielding a significant improvement over the unconditional separation framework. Recent works [18,19] have utilized conditional embeddings for each music class in order to boost the performance of a deep attractor-network [20] for music separation.…”

Section: Introductionmentioning

confidence: 99%

Improving Universal Sound Separation Using Sound Classification

Tzinis

Wisdom

Hershey

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Deep learning approaches have recently achieved impressive performance on both audio source separation and sound classification. Most audio source separation approaches focus only on separating sources belonging to a restricted domain of source classes, such as speech and music. However, recent work has demonstrated the possibility of "universal sound separation", which aims to separate acoustic sources from an open domain, regardless of their class. In this paper, we utilize the semantic information learned by sound classifier networks trained on a vast amount of diverse sounds to improve universal sound separation. In particular, we show that semantic embeddings extracted from a sound classifier can be used to condition a separation network, providing it with useful additional information. This approach is especially useful in an iterative setup, where source estimates from an initial separation stage and their corresponding classifier-derived embeddings are fed to a second separation network. By performing a thorough hyperparameter search consisting of over a thousand experiments, we find that classifier embeddings from clean sources provide nearly one dB of SNR gain, and our best iterative models achieve a significant fraction of this oracle performance, establishing a new state-of-the-art for universal sound separation.

show abstract

Speaker Diarization Using Deep Recurrent Convolutional Neural Networks for Speaker Embeddings

Cited by 23 publications

References 35 publications

Robust Speaker Recognition Using Unsupervised Adversarial Invariance

Robust Speaker Recognition Using Unsupervised Adversarial Invariance

Capturing Suprasegmental Features of a Voice with RNNs for Improved Speaker Clustering

Improving Universal Sound Separation Using Sound Classification

Contact Info

Product

Resources

About