Hong-Goo Kang

Hyeon-Kyeong Shin, Hyewon Han, Doyeon Kim et al.

In this paper, we propose a novel end-to-end user-defined keyword spotting method that utilizes linguistically corresponding patterns between speech and text sequences. Unlike previous approaches requiring speech keyword enrollment, our method compares input queries with an enrolled text keyword sequence. To place the audio and text representations within a common latent space, we adopt an attention-based cross-modal matching approach that is trained in an end-to-end manner with monotonic matching loss and keyword classification loss. We also utilize a de-noising loss for the acoustic embedding network to improve robustness in noisy environments. Additionally, we introduce the LibriPhrase dataset, a new short-phrase dataset based on LibriSpeech for efficiently training keyword spotting models. Our proposed method achieves competitive results on various evaluation sets compared to other single-modal and cross-modal baselines.

Hyungchan Yoon, Changhwan Kim, Eunwoo Song et al.

For personalized speech generation, a neural text-to-speech (TTS) model must be successfully implemented with limited data from a target speaker. To this end, the baseline TTS model needs to be amply generalized to out-of-domain data (i.e., target speaker's speech). However, approaches to address this out-of-domain generalization problem in TTS have yet to be thoroughly studied. In this work, we propose an effective pruning method for a transformer known as sparse attention, to improve the TTS model's generalization abilities. In particular, we prune off redundant connections from self-attention layers whose attention weights are below the threshold. To flexibly determine the pruning strength for searching optimal degree of generalization, we also propose a new differentiable pruning method that allows the model to automatically learn the thresholds. Evaluations on zero-shot multi-speaker TTS verify the effectiveness of our method in terms of voice quality and speaker similarity.

Sangshin Oh, Seyun Um, Hong-Goo Kang

The Gumbel-softmax distribution, or Concrete distribution, is often used to relax the discrete characteristics of a categorical distribution and enable back-propagation through differentiable reparameterization. Although it reliably yields low variance gradients, it still relies on a stochastic sampling process for optimization. In this work, we present a relaxed categorical analytic bound (ReCAB), a novel divergence-like metric which corresponds to the upper bound of the Kullback-Leibler divergence (KLD) of a relaxed categorical distribution. The proposed metric is easy to implement because it has a closed form solution, and empirical results show that it is close to the actual KLD. Along with this new metric, we propose a relaxed categorical analytic bound variational autoencoder (ReCAB-VAE) that successfully models both continuous and relaxed discrete latent representations. We implement an emotional text-to-speech synthesis system based on the proposed framework, and show that the proposed system flexibly and stably controls emotion expressions with better speech quality compared to baselines that use stochastic estimation or categorical distribution approximation.

Hejung Yang, Hong-Goo Kang

Large, pre-trained representation models trained using self-supervised learning have gained popularity in various fields of machine learning because they are able to extract high-quality salient features from input data. As such, they have been frequently used as base networks for various pattern classification tasks such as speech recognition. However, not much research has been conducted on applying these types of models to the field of speech signal generation. In this paper, we investigate the feasibility of using pre-trained speech representation models for a downstream speech enhancement task. To alleviate mismatches between the input features of the pre-trained model and the target enhancement model, we adopt a novel feature normalization technique to smoothly link these modules together. Our proposed method enables significant improvements in speech quality compared to baselines when combined with various types of pre-trained speech models.

Yeona Hong, Woo-Jin Chung, Hong-Goo Kang

As Deep Neural Networks (DNNs) rapidly advance in various fields, including speech verification, they typically involve high computational costs and substantial memory consumption, which can be challenging to manage on mobile systems. Quantization of deep models offers a means to reduce both computational and memory expenses. Our research proposes an optimization framework for the quantization of the speaker verification model. By analyzing performance changes and model size reductions in each layer of a pre-trained speaker verification model, we have effectively minimized performance degradation while significantly reducing the model size. Our quantization algorithm is the first attempt to maintain the performance of the state-of-the-art pre-trained speaker verification model, ECAPATDNN, while significantly compressing its model size. Overall, our quantization approach resulted in reducing the model size by half, with an increase in EER limited to 0.07%.

Sangmin Lee, Woojin Chung, Seyun Um et al.

Code-switching (CS), the alternation between two or more languages within a single speaker's utterances, is common in real-world conversations and poses significant challenges for multilingual speech technology. However, systems capable of handling this phenomenon remain underexplored, primarily due to the scarcity of suitable datasets. To resolve this issue, we propose Universal Code-Mixer (UniCoM), a novel pipeline for generating high-quality, natural CS samples without altering sentence semantics. Our approach utilizes an algorithm we call Substituting WORDs with Synonyms (SWORDS), which generates CS speech by replacing selected words with their translations while considering their parts of speech. Using UniCoM, we construct Code-Switching FLEURS (CS-FLEURS), a multilingual CS corpus designed for automatic speech recognition (ASR) and speech-to-text translation (S2TT). Experimental results show that CS-FLEURS achieves high intelligibility and naturalness, performing comparably to existing datasets on both objective and subjective metrics. We expect our approach to advance CS speech technology and enable more inclusive multilingual systems.

Woongjib Choi, Sangmin Lee, Hyungseob Lim et al.

In this paper, we present a vocoder-free framework for audio super-resolution that employs a flow matching generative model to capture the conditional distribution of complex-valued spectral coefficients. Unlike conventional two-stage diffusion-based approaches that predict a mel-spectrogram and then rely on a pre-trained neural vocoder to synthesize waveforms, our method directly reconstructs waveforms via the inverse Short-Time Fourier Transform (iSTFT), thereby eliminating the dependence on a separate vocoder. This design not only simplifies end-to-end optimization but also overcomes a critical bottleneck of two-stage pipelines, where the final audio quality is fundamentally constrained by vocoder performance. Experiments show that our model consistently produces high-fidelity 48 kHz audio across diverse upsampling factors, achieving state-of-the-art performance on both speech and general audio datasets.

Sangmin Lee, Woongjib Choi, Jihyun Kim et al.

In this paper, we present a neural spoken language diarization model that supports an unconstrained span of languages within a single framework. Our approach integrates a learnable query-based architecture grounded in multilingual awareness, with large-scale pretraining on simulated code-switching data. By jointly leveraging these two components, our method overcomes the limitations of conventional approaches in data scarcity and architecture optimization, and generalizes effectively to real-world multilingual settings across diverse environments. Experimental results demonstrate that our approach achieves state-of-the-art performance on several language diarization benchmarks, with a relative performance improvement of 23% to 52% over previous methods. We believe that this work not only advances research in language diarization but also establishes a foundational framework for code-switching speech technologies.

Yeona Hong, Hyewon Han, Woo-jin Chung et al.

In this paper, we propose StableQuant, a novel adaptive post-training quantization (PTQ) algorithm for widely used speech foundation models (SFMs). While PTQ has been successfully employed for compressing large language models (LLMs) due to its ability to bypass additional fine-tuning, directly applying these techniques to SFMs may not yield optimal results, as SFMs utilize distinct network architecture for feature extraction. StableQuant demonstrates optimal quantization performance regardless of the network architecture type, as it adaptively determines the quantization range for each layer by analyzing both the scale distributions and overall performance. We evaluate our algorithm on two SFMs, HuBERT and wav2vec2.0, for an automatic speech recognition (ASR) task, and achieve superior performance compared to traditional PTQ methods. StableQuant successfully reduces the sizes of SFM models to a quarter and doubles the inference speed while limiting the word error rate (WER) performance drop to less than 0.3% with 8-bit quantization.

Sangmin Lee, Woo-Jin Chung, Hong-Goo Kang

Building a universal multilingual automatic speech recognition (ASR) model that performs equitably across languages has long been a challenge due to its inherent difficulties. To address this task we introduce a Language-Agnostic Multilingual ASR pipeline through orthography Unification and language-specific Transliteration (LAMA-UT). LAMA-UT operates without any language-specific modules while matching the performance of state-of-the-art models trained on a minimal amount of data. Our pipeline consists of two key steps. First, we utilize a universal transcription generator to unify orthographic features into Romanized form and capture common phonetic characteristics across diverse languages. Second, we utilize a universal converter to transform these universal transcriptions into language-specific ones. In experiments, we demonstrate the effectiveness of our proposed method leveraging universal transcriptions for massively multilingual ASR. Our pipeline achieves a relative error reduction rate of 45% when compared to Whisper and performs comparably to MMS, despite being trained on only 0.1% of Whisper's training data. Furthermore, our pipeline does not rely on any language-specific modules. However, it performs on par with zero-shot ASR approaches which utilize additional language-specific lexicons and language models. We expect this framework to serve as a cornerstone for flexible multilingual ASR systems that are generalizable even to unseen languages.

Doyeon Kim, Hyewon Han, Hyeon-Kyeong Shin et al.

Modern neural speech enhancement models usually include various forms of phase information in their training loss terms, either explicitly or implicitly. However, these loss terms are typically designed to reduce the distortion of phase spectrum values at specific frequencies, which ensures they do not significantly affect the quality of the enhanced speech. In this paper, we propose an effective phase reconstruction strategy for neural speech enhancement that can operate in noisy environments. Specifically, we introduce a phase continuity loss that considers relative phase variations across the time and frequency axes. By including this phase continuity loss in a state-of-the-art neural speech enhancement system trained with reconstruction loss and a number of magnitude spectral losses, we show that our proposed method further improves the quality of enhanced speech signals over the baseline, especially when training is done jointly with a magnitude spectrum loss.

Jee-weon Jung, Hemlata Tak, Hye-jin Shim et al.

ASV (automatic speaker verification) systems are intrinsically required to reject both non-target (e.g., voice uttered by different speaker) and spoofed (e.g., synthesised or converted) inputs. However, there is little consideration for how ASV systems themselves should be adapted when they are expected to encounter spoofing attacks, nor when they operate in tandem with CMs (spoofing countermeasures), much less how both systems should be jointly optimised. The goal of the first SASV (spoofing-aware speaker verification) challenge, a special sesscion in ISCA INTERSPEECH 2022, is to promote development of integrated systems that can perform ASV and CM simultaneously.

Se-Yun Um, Jihyun Kim, Jihyun Lee et al.

In this paper, we propose a multi-speaker face-to-speech waveform generation model that also works for unseen speaker conditions. Using a generative adversarial network (GAN) with linguistic and speaker characteristic features as auxiliary conditions, our method directly converts face images into speech waveforms under an end-to-end training framework. The linguistic features are extracted from lip movements using a lip-reading model, and the speaker characteristic features are predicted from face images using cross-modal learning with a pre-trained acoustic model. Since these two features are uncorrelated and controlled independently, we can flexibly synthesize speech waveforms whose speaker characteristics vary depending on the input face images. We show the superiority of our proposed model over conventional methods in terms of objective and subjective evaluation results. Specifically, we evaluate the performances of linguistic features by measuring their accuracy on an automatic speech recognition task. In addition, we estimate speaker and gender similarity for multi-speaker and unseen conditions, respectively. We also evaluate the aturalness of the synthesized speech waveforms using a mean opinion score (MOS) test and non-intrusive objective speech quality assessment (NISQA).The demo samples of the proposed and other models are available at https://sam-0927.github.io/

Jiyoung Lee, Soo-Whan Chung, Sunok Kim et al.

In this paper, we address the problem of separating individual speech signals from videos using audio-visual neural processing. Most conventional approaches utilize frame-wise matching criteria to extract shared information between co-occurring audio and video. Thus, their performance heavily depends on the accuracy of audio-visual synchronization and the effectiveness of their representations. To overcome the frame discontinuity problem between two modalities due to transmission delay mismatch or jitter, we propose a cross-modal affinity network (CaffNet) that learns global correspondence as well as locally-varying affinities between audio and visual streams. Given that the global term provides stability over a temporal sequence at the utterance-level, this resolves the label permutation problem characterized by inconsistent assignments. By extending the proposed cross-modal affinity on the complex network, we further improve the separation performance in the complex spectral domain. Experimental results verify that the proposed methods outperform conventional ones on various datasets, demonstrating their advantages in real-world scenarios.

Hyewon Han, Soo-Whan Chung, Hong-Goo Kang

Many approaches can derive information about a single speaker's identity from the speech by learning to recognize consistent characteristics of acoustic parameters. However, it is challenging to determine identity information when there are multiple concurrent speakers in a given signal. In this paper, we propose a novel deep speaker representation strategy that can reliably extract multiple speaker identities from an overlapped speech. We design a network that can extract a high-level embedding that contains information about each speaker's identity from a given mixture. Unlike conventional approaches that need reference acoustic features for training, our proposed algorithm only requires the speaker identity labels of the overlapped speech segments. We demonstrate the effectiveness and usefulness of our algorithm in a speaker verification task and a speech separation system conditioned on the target speaker embeddings obtained through the proposed method.

Yoohwan Kwon, Soo-Whan Chung, Hong-Goo Kang

In this paper, we propose an effective training strategy to ex-tract robust speaker representations from a speech signal. Oneof the key challenges in speaker recognition tasks is to learnlatent representations or embeddings containing solely speakercharacteristic information in order to be robust in terms of intra-speaker variations. By modifying the network architecture togenerate both speaker-related and speaker-unrelated representa-tions, we exploit a learning criterion which minimizes the mu-tual information between these disentangled embeddings. Wealso introduce an identity change loss criterion which utilizes areconstruction error to different utterances spoken by the samespeaker. Since the proposed criteria reduce the variation ofspeaker characteristics caused by changes in background envi-ronment or spoken content, the resulting embeddings of eachspeaker become more consistent. The effectiveness of the pro-posed method is demonstrated through two tasks; disentangle-ment performance, and improvement of speaker recognition ac-curacy compared to the baseline model on a benchmark dataset,VoxCeleb1. Ablation studies also show the impact of each cri-terion on overall performance.

Soo-Whan Chung, Soyeon Choe, Joon Son Chung et al.

The objective of this paper is to separate a target speaker's speech from a mixture of two speakers using a deep audio-visual speech separation network. Unlike previous works that used lip movement on video clips or pre-enrolled speaker information as an auxiliary conditional feature, we use a single face image of the target speaker. In this task, the conditional feature is obtained from facial appearance in cross-modal biometric task, where audio and visual identity representations are shared in latent space. Learnt identities from facial images enforce the network to isolate matched speakers and extract the voices from mixed speech. It solves the permutation problem caused by swapped channel outputs, frequently occurred in speech separation tasks. The proposed method is far more practical than video-based speech separation since user profile images are readily available on many platforms. Also, unlike speaker-aware separation methods, it is applicable on separation with unseen speakers who have never been enrolled before. We show strong qualitative and quantitative results on challenging real-world examples.

Se-Yun Um, Sangshin Oh, Kyungguen Byun et al.

This paper proposes an effective emotion control method for an end-to-end text-to-speech (TTS) system. To flexibly control the distinct characteristic of a target emotion category, it is essential to determine embedding vectors representing the TTS input. We introduce an inter-to-intra emotional distance ratio algorithm to the embedding vectors that can minimize the distance to the target emotion category while maximizing its distance to the other emotion categories. To further enhance the expressiveness of a target speech, we also introduce an effective interpolation technique that enables the intensity of a target emotion to be gradually changed to that of neutral speech. Subjective evaluation results in terms of emotional expressiveness and controllability show the superiority of the proposed algorithm to the conventional methods.

Min-Jae Hwang, Hong-Goo Kang

In this paper, we propose a deep learning (DL)-based parameter enhancement method for a mixed excitation linear prediction (MELP) speech codec in noisy communication environment. Unlike conventional speech enhancement modules that are designed to obtain clean speech signal by removing noise components before speech codec processing, the proposed method directly enhances codec parameters on either the encoder or decoder side. As the proposed method has been implemented by a small network without any additional processes required in conventional enhancement systems, e.g., time-frequency (T-F) analysis/synthesis modules, its computational complexity is very low. By enhancing the noise-corrupted codec parameters with the proposed DL framework, we achieved an enhancement system that is much simpler and faster than conventional T-F mask-based speech enhancement methods, while the quality of its performance remains similar.

Ohsung Kwon, Eunwoo Song, Jae-Min Kim et al.

In this paper, we propose a high-quality generative text-to-speech (TTS) system using an effective spectrum and excitation estimation method. Our previous research verified the effectiveness of the ExcitNet-based speech generation model in a parametric TTS framework. However, the challenge remains to build a high-quality speech synthesis system because auxiliary conditional features estimated by a simple deep neural network often contain large prediction errors, and the errors are inevitably propagated throughout the autoregressive generation process of the ExcitNet vocoder. To generate more natural speech signals, we exploited a sequence-to-sequence (seq2seq) acoustic model with an attention-based generative network (e.g., Tacotron 2) to estimate the condition parameters of the ExcitNet vocoder. Because the seq2seq acoustic model accurately estimates spectral parameters, and because the ExcitNet model effectively generates the corresponding time-domain excitation signals, combining these two models can synthesize natural speech signals. Furthermore, we verified the merit of the proposed method in producing expressive speech segments by adopting a global style token-based emotion embedding method. The experimental results confirmed that the proposed system significantly outperforms the systems with a similarly configured conventional WaveNet vocoder and our best prior parametric TTS counterpart.

Soyeon Choe, Soo-Whan Chung, Youna Ji et al.

Deep clustering is a deep neural network-based speech separation algorithm that first trains the mixed component of signals with high-dimensional embeddings, and then uses a clustering algorithm to separate each mixture of sources. In this paper, we extend the baseline criterion of deep clustering with an additional regularization term to further improve the overall performance. This term plays a role in assigning a condition to the embeddings such that it gives less correlation to each embedding dimension, leading to better decomposition of the spectral bins. The regularization term helps to mitigate the unavoidable permutation problem in the conventional deep clustering method, which enables to bring better clustering through the formation of optimal embeddings. We evaluate the results by varying embedding dimension, signal-to-interference ratio (SIR), and gender dependency. The performance comparison with the source separation measurement metric, i.e. signal-to-distortion ratio (SDR), confirms that the proposed method outperforms the conventional deep clustering method.

Min-Jae Hwang, Frank Soong, Eunwoo Song et al.

We propose a linear prediction (LP)-based waveform generation method via WaveNet vocoding framework. A WaveNet-based neural vocoder has significantly improved the quality of parametric text-to-speech (TTS) systems. However, it is challenging to effectively train the neural vocoder when the target database contains massive amount of acoustical information such as prosody, style or expressiveness. As a solution, the approaches that only generate the vocal source component by a neural vocoder have been proposed. However, they tend to generate synthetic noise because the vocal source component is independently handled without considering the entire speech production process; where it is inevitable to come up with a mismatch between vocal source and vocal tract filter. To address this problem, we propose an LP-WaveNet vocoder, where the complicated interactions between vocal source and vocal tract components are jointly trained within a mixture density network-based WaveNet model. The experimental results verify that the proposed system outperforms the conventional WaveNet vocoders both objectively and subjectively. In particular, the proposed method achieves 4.47 MOS within the TTS framework.

Eunwoo Song, Kyungguen Byun, Hong-Goo Kang

This paper proposes a WaveNet-based neural excitation model (ExcitNet) for statistical parametric speech synthesis systems. Conventional WaveNet-based neural vocoding systems significantly improve the perceptual quality of synthesized speech by statistically generating a time sequence of speech waveforms through an auto-regressive framework. However, they often suffer from noisy outputs because of the difficulties in capturing the complicated time-varying nature of speech signals. To improve modeling efficiency, the proposed ExcitNet vocoder employs an adaptive inverse filter to decouple spectral components from the speech signal. The residual component, i.e. excitation signal, is then trained and generated within the WaveNet framework. In this way, the quality of the synthesized speech signal can be further improved since the spectral component is well represented by a deep learning framework and, moreover, the residual component is efficiently generated by the WaveNet framework. Experimental results show that the proposed ExcitNet vocoder, trained both speaker-dependently and speaker-independently, outperforms traditional linear prediction vocoders and similarly configured conventional WaveNet vocoders.

Eunwoo Song, Jin-Seob Kim, Kyungguen Byun et al.

This paper proposes speaker-adaptive neural vocoders for parametric text-to-speech (TTS) systems. Recently proposed WaveNet-based neural vocoding systems successfully generate a time sequence of speech signal with an autoregressive framework. However, it remains a challenge to synthesize high-quality speech when the amount of a target speaker's training data is insufficient. To generate more natural speech signals with the constraint of limited training data, we propose a speaker adaptation task with an effective variation of neural vocoding models. In the proposed method, a speaker-independent training method is applied to capture universal attributes embedded in multiple speakers, and the trained model is then optimized to represent the specific characteristics of the target speaker. Experimental results verify that the proposed TTS systems with speaker-adaptive neural vocoders outperform those with traditional source-filter model-based vocoders and those with WaveNet vocoders, trained either speaker-dependently or speaker-independently. In particular, our TTS system achieves 3.80 and 3.77 MOS for the Korean male and Korean female speakers, respectively, even though we use only ten minutes' speech corpus for training the model.

Soo-Whan Chung, Joon Son Chung, Hong-Goo Kang

This paper proposes a new strategy for learning powerful cross-modal embeddings for audio-to-video synchronization. Here, we set up the problem as one of cross-modal retrieval, where the objective is to find the most relevant audio segment given a short video clip. The method builds on the recent advances in learning representations from cross-modal self-supervision. The main contributions of this paper are as follows: (1) we propose a new learning strategy where the embeddings are learnt via a multi-way matching problem, as opposed to a binary classification (matching or non-matching) problem as proposed by recent papers; (2) we demonstrate that performance of this method far exceeds the existing baselines on the synchronization task; (3) we use the learnt embeddings for visual speech recognition in self-supervision, and show that the performance matches the representations learnt end-to-end in a fully-supervised manner.