arXiv subjects

Hoirin Kim

Publications and source records attributed to Hoirin Kim.

35 records · Page 2Linked to original sources

arXiv2020

Supervised attention for speaker recognition

The recently proposed self-attentive pooling (SAP) has shown good performance in several speaker recognition systems. In SAP systems, the context vector is trained end-to-end together with the feature extractor, where the role of context vector is to select the most discriminative frames for speaker recognition. However, the SAP underperforms compared to the temporal average pooling (TAP) baseline in some settings, which implies that the attention is not learnt effectively in end-to-end training. To tackle this problem, we introduce strategies for training the attention mechanism in a supervised manner, which learns the context vector using classified samples. With our proposed methods, context vector can be boosted to select the most informative frames. We show that our method outperforms existing methods in various experimental settings including short utterance speaker recognition, and achieves competitive performance over the existing baselines on the VoxCeleb datasets.

cs.SD↗

arXiv2020

Neural MOS Prediction for Synthesized Speech Using Multi-Task Learning With Spoofing Detection and Spoofing Type Classification

Several studies have proposed deep-learning-based models to predict the mean opinion score (MOS) of synthesized speech, showing the possibility of replacing human raters. However, inter- and intra-rater variability in MOSs makes it hard to ensure the high performance of the models. In this paper, we propose a multi-task learning (MTL) method to improve the performance of a MOS prediction model using the following two auxiliary tasks: spoofing detection (SD) and spoofing type classification (STC). Besides, we use the focal loss to maximize the synergy between SD and STC for MOS prediction. Experiments using the MOS evaluation results of the Voice Conversion Challenge 2018 show that proposed MTL with two auxiliary tasks improves MOS prediction. Our proposed model achieves up to 11.6% relative improvement in performance over the baseline model.

eess.AS↗

arXiv2020

A Unified Deep Learning Framework for Short-Duration Speaker Verification in Adverse Environments

Speaker verification (SV) has recently attracted considerable research interest due to the growing popularity of virtual assistants. At the same time, there is an increasing requirement for an SV system: it should be robust to short speech segments, especially in noisy and reverberant environments. In this paper, we consider one more important requirement for practical applications: the system should be robust to an audio stream containing long non-speech segments, where a voice activity detection (VAD) is not applied. To meet these two requirements, we introduce feature pyramid module (FPM)-based multi-scale aggregation (MSA) and self-adaptive soft VAD (SAS-VAD). We present the FPM-based MSA to deal with short speech segments in noisy and reverberant environments. Also, we use the SAS-VAD to increase the robustness to long non-speech segments. To further improve the robustness to acoustic distortions (i.e., noise and reverberation), we apply a masking-based speech enhancement (SE) method. We combine SV, VAD, and SE models in a unified deep learning framework and jointly train the entire network in an end-to-end manner. To the best of our knowledge, this is the first work combining these three models in a deep learning framework. We conduct experiments on Korean indoor (KID) and VoxCeleb datasets, which are corrupted by noise and reverberation. The results show that the proposed method is effective for SV in the challenging conditions and performs better than the baseline i-vector and deep speaker embedding systems.

eess.AS↗

arXiv2020

Dual Attention in Time and Frequency Domain for Voice Activity Detection

Voice activity detection (VAD) is a challenging task in low signal-to-noise ratio (SNR) environment, especially in non-stationary noise. To deal with this issue, we propose a novel attention module that can be integrated in Long Short-Term Memory (LSTM). Our proposed attention module refines each LSTM layer's hidden states so as to make it possible to adaptively focus on both time and frequency domain. Experiments are conducted on various noisy conditions using Aurora 4 database. Our proposed method obtains the 95.58 % area under the ROC curve (AUC), achieving 22.05 % relative improvement compared to baseline, with only 2.44 % increase in the number of parameters. Besides, we utilize focal loss for alleviating the performance degradation caused by imbalance between speech and non-speech sections in training sets. The results show that the focal loss can improve the performance in various imbalance situations compared to the cross entropy loss, a commonly used loss function in VAD.

eess.AS↗

arXiv2020

Meta-Learning for Short Utterance Speaker Recognition with Imbalance Length Pairs

In practical settings, a speaker recognition system needs to identify a speaker given a short utterance, while the enrollment utterance may be relatively long. However, existing speaker recognition models perform poorly with such short utterances. To solve this problem, we introduce a meta-learning framework for imbalance length pairs. Specifically, we use a Prototypical Networks and train it with a support set of long utterances and a query set of short utterances of varying lengths. Further, since optimizing only for the classes in the given episode may be insufficient for learning discriminative embeddings for unseen classes, we additionally enforce the model to classify both the support and the query set against the entire set of classes in the training set. By combining these two learning schemes, our model outperforms existing state-of-the-art speaker verification models learned with a standard supervised learning framework on short utterance (1-2 seconds) on the VoxCeleb datasets. We also validate our proposed model for unseen speaker identification, on which it also achieves significant performance gains over the existing approaches. The codes are available at https://github.com/seongmin-kye/meta-SR.

eess.AS↗

arXiv2020

Deep MOS Predictor for Synthetic Speech Using Cluster-Based Modeling

While deep learning has made impressive progress in speech synthesis and voice conversion, the assessment of the synthesized speech is still carried out by human participants. Several recent papers have proposed deep-learning-based assessment models and shown the potential to automate the speech quality assessment. To improve the previously proposed assessment model, MOSNet, we propose three models using cluster-based modeling methods: using a global quality token (GQT) layer, using an Encoding Layer, and using both of them. We perform experiments using the evaluation results of the Voice Conversion Challenge 2018 to predict the mean opinion score of synthesized speech and similarity score between synthesized speech and reference speech. The results show that the GQT layer helps to predict human assessment better by automatically learning the useful quality tokens for the task and that the Encoding Layer helps to utilize frame-level scores more precisely.

eess.AS↗

arXiv2020

Multi-Task Network for Noise-Robust Keyword Spotting and Speaker Verification using CTC-based Soft VAD and Global Query Attention

Keyword spotting (KWS) and speaker verification (SV) have been studied independently although it is known that acoustic and speaker domains are complementary. In this paper, we propose a multi-task network that performs KWS and SV simultaneously to fully utilize the interrelated domain information. The multi-task network tightly combines sub-networks aiming at performance improvement in challenging conditions such as noisy environments, open-vocabulary KWS, and short-duration SV, by introducing novel techniques of connectionist temporal classification (CTC)-based soft voice activity detection (VAD) and global query attention. Frame-level acoustic and speaker information is integrated with phonetically originated weights so that forms a word-level global representation. Then it is used for the aggregation of feature vectors to generate discriminative embeddings. Our proposed approach shows 4.06% and 26.71% relative improvements in equal error rate (EER) compared to the baselines for both tasks. We also present a visualization example and results of ablation experiments.

eess.AS↗

arXiv2020

Improving Multi-Scale Aggregation Using Feature Pyramid Module for Robust Speaker Verification of Variable-Duration Utterances

Currently, the most widely used approach for speaker verification is the deep speaker embedding learning. In this approach, we obtain a speaker embedding vector by pooling single-scale features that are extracted from the last layer of a speaker feature extractor. Multi-scale aggregation (MSA), which utilizes multi-scale features from different layers of the feature extractor, has recently been introduced and shows superior performance for variable-duration utterances. To increase the robustness dealing with utterances of arbitrary duration, this paper improves the MSA by using a feature pyramid module. The module enhances speaker-discriminative information of features from multiple layers via a top-down pathway and lateral connections. We extract speaker embeddings using the enhanced features that contain rich speaker information with different time scales. Experiments on the VoxCeleb dataset show that the proposed module improves previous MSA methods with a smaller number of parameters. It also achieves better performance than state-of-the-art approaches for both short and long utterances.

eess.AS↗

arXiv2020

Meta-Learned Confidence for Few-shot Learning

Transductive inference is an effective means of tackling the data deficiency problem in few-shot learning settings. A popular transductive inference technique for few-shot metric-based approaches, is to update the prototype of each class with the mean of the most confident query examples, or confidence-weighted average of all the query samples. However, a caveat here is that the model confidence may be unreliable, which may lead to incorrect predictions. To tackle this issue, we propose to meta-learn the confidence for each query sample, to assign optimal weights to unlabeled queries such that they improve the model's transductive inference performance on unseen tasks. We achieve this by meta-learning an input-adaptive distance metric over a task distribution under various model and data perturbations, which will enforce consistency on the model predictions under diverse uncertainties for unseen tasks. Moreover, we additionally suggest a regularization which explicitly enforces the consistency on the predictions across the different dimensions of a high-dimensional embedding vector. We validate our few-shot learning model with meta-learned confidence on four benchmark datasets, on which it largely outperforms strong recent baselines and obtains new state-of-the-art results. Further application on semi-supervised few-shot learning tasks also yields significant performance improvements over the baselines. The source code of our algorithm is available at https://github.com/seongmin-kye/MCT.

cs.LG↗

arXiv2020

Non-parallel voice conversion based on source-to-target direct mapping

Recent works of utilizing phonetic posteriograms (PPGs) for non-parallel voice conversion have significantly increased the usability of voice conversion since the source and target DBs are no longer required for matching contents. In this approach, the PPGs are used as the linguistic bridge between source and target speaker features. However, this PPG-based non-parallel voice conversion has some limitation that it needs two cascading networks at conversion time, making it less suitable for real-time applications and vulnerable to source speaker intelligibility at conversion stage. To address this limitation, we propose a new non-parallel voice conversion technique that employs a single neural network for direct source-to-target voice parameter mapping. With this single network structure, the proposed approach can reduce both conversion time and number of network parameters, which can be especially important factors in embedded or real-time environments. Additionally, it improves the quality of voice conversion by skipping the phone recognizer at conversion stage. It can effectively prevent possible loss of phonetic information the PPG-based indirect method suffers. Experiments show that our approach reduces number of network parameters and conversion time by 41.9% and 44.5%, respectively, with improved voice similarity over the original PPG-based method.

eess.AS↗

arXiv2020

Neural voice cloning with a few low-quality samples

In this paper, we explore the possibility of speech synthesis from low quality found data using only limited number of samples of target speaker. We try to extract only the speaker embedding from found data of target speaker unlike previous works which tries to train the entire text-to-speech system on found data. Also, the two speaker mimicking approaches which are adaptation and speaker-encoder-based are applied on newly released LibriTTS dataset and previously released VCTK corpus to examine the impact of speaker variety on clarity and target-speaker-similarity .

eess.AS↗

arXiv2020

Domain-adversarial training of multi-speaker TTS

Multi-speaker TTS has to learn both linguistic embedding and text embedding to generate speech of desired linguistic content in desired voice. However, it is unclear which characteristic of speech results from speaker and which part from linguistic content. In this paper, text embedding is forced to unlearn speaker dependent characteristic using gradient reversal layer to auxiliary speaker classifier that we introduce. We train a speaker classifier using angular margin softmax loss. In subjective evaluation, it is shown that the adversarial training of text embedding for unilingual multi-speaker TTS results in 39.9% improvement on similarity MOS and 40.1% improvement on naturalness MOS.

eess.AS↗

arXiv2020

Pitchtron: Towards audiobook generation from ordinary people's voices

In this paper, we explore prosody transfer for audiobook generation under rather realistic condition where training DB is plain audio mostly from multiple ordinary people and reference audio given during inference is from professional and richer in prosody than training DB. To be specific, we explore transferring Korean dialects and emotive speech even though training set is mostly composed of standard and neutral Korean. We found that under this setting, original global style token method generates undesirable glitches in pitch, energy and pause length. To deal with this issue, we propose two models, hard and soft pitchtron and release the toolkit and corpus that we have developed. Hard pitchtron uses pitch as input to the decoder while soft pitchtron uses pitch as input to the prosody encoder. We verify the effectiveness of proposed models with objective and subjective tests. AXY score over GST is 2.01 and 1.14 for hard pitchtron and soft pitchtron respectively.

eess.AS↗

arXiv2019

Additional Shared Decoder on Siamese Multi-view Encoders for Learning Acoustic Word Embeddings

Acoustic word embeddings --- fixed-dimensional vector representations of arbitrary-length words --- have attracted increasing interest in query-by-example spoken term detection. Recently, on the fact that the orthography of text labels partly reflects the phonetic similarity between the words' pronunciation, a multi-view approach has been introduced that jointly learns acoustic and text embeddings. It showed that it is possible to learn discriminative embeddings by designing the objective which takes text labels as well as word segments. In this paper, we propose a network architecture that expands the multi-view approach by combining the Siamese multi-view encoders with a shared decoder network to maximize the effect of the relationship between acoustic and text embeddings in embedding space. Discriminatively trained with multi-view triplet loss and decoding loss, our proposed approach achieves better performance on acoustic word discrimination task with the WSJ dataset, resulting in 11.1% relative improvement in average precision. We also present experimental results on cross-view word discrimination and word level speech recognition tasks.

eess.AS↗

arXiv2019

Self-Adaptive Soft Voice Activity Detection using Deep Neural Networks for Robust Speaker Verification

Voice activity detection (VAD), which classifies frames as speech or non-speech, is an important module in many speech applications including speaker verification. In this paper, we propose a novel method, called self-adaptive soft VAD, to incorporate a deep neural network (DNN)-based VAD into a deep speaker embedding system. The proposed method is a combination of the following two approaches. The first approach is soft VAD, which performs a soft selection of frame-level features extracted from a speaker feature extractor. The frame-level features are weighted by their corresponding speech posteriors estimated from the DNN-based VAD, and then aggregated to generate a speaker embedding. The second approach is self-adaptive VAD, which fine-tunes the pre-trained VAD on the speaker verification data to reduce the domain mismatch. Here, we introduce two unsupervised domain adaptation (DA) schemes, namely speech posterior-based DA (SP-DA) and joint learning-based DA (JL-DA). Experiments on a Korean speech database demonstrate that the verification performance is improved significantly in real-world environments by using self-adaptive soft VAD.

eess.AS↗

arXiv2019

Spatial Pyramid Encoding with Convex Length Normalization for Text-Independent Speaker Verification

In this paper, we propose a new pooling method called spatial pyramid encoding (SPE) to generate speaker embeddings for text-independent speaker verification. We first partition the output feature maps from a deep residual network (ResNet) into increasingly fine sub-regions and extract speaker embeddings from each sub-region through a learnable dictionary encoding layer. These embeddings are concatenated to obtain the final speaker representation. The SPE layer not only generates a fixed-dimensional speaker embedding for a variable-length speech segment, but also aggregates the information of feature distribution from multi-level temporal bins. Furthermore, we apply deep length normalization by augmenting the loss function with ring loss. By applying ring loss, the network gradually learns to normalize the speaker embeddings using model weights themselves while preserving convexity, leading to more robust speaker embeddings. Experiments on the VoxCeleb1 dataset show that the proposed system using the SPE layer and ring loss-based deep length normalization outperforms both i-vector and d-vector baselines.

eess.AS↗

arXiv2018

Learning acoustic word embeddings with phonetically associated triplet network

Previous researches on acoustic word embeddings used in query-by-example spoken term detection have shown remarkable performance improvements when using a triplet network. However, the triplet network is trained using only a limited information about acoustic similarity between words. In this paper, we propose a novel architecture, phonetically associated triplet network (PATN), which aims at increasing discriminative power of acoustic word embeddings by utilizing phonetic information as well as word identity. The proposed model is learned to minimize a combined loss function that was made by introducing a cross entropy loss to the lower layer of LSTM-based triplet network. We observed that the proposed method performs significantly better than the baseline triplet network on a word discrimination task with the WSJ dataset resulting in over 20% relative improvement in recall rate at 1.0 false alarm per hour. Finally, we examined the generalization ability by conducting the out-of-domain test on the RM dataset.

eess.AS↗