arXiv subjects

Zhizheng Wu

Publications and source records attributed to Zhizheng Wu.

At least 73 records · Page 4Linked to original sources

arXiv2025

TASLA: Text-Aligned Speech Tokens with Multiple Layer-Aggregation

We propose Text-Aligned Speech Tokens with Multiple Layer-Aggregation (TASLA), which is a text-aligned speech tokenization framework that aims to address the problem that under a low-frame-rate and text-aligned regime, single-source speech tokens may lose acoustic details during reconstruction. On the other hand, this paper further explains how different encoder layers collaborate to capture comprehensive acoustic features for tokenization. Previous work, TASTE, proposed the text-aligned speech tokenization framework, which is a LM-friendly architecture, but struggles to capture acoustic details. We address this trade-off with two components: Multi-Layer Dynamic Attention (MLDA), which lets each text position adaptively mix shallow/deep features from a frozen speech encoder, and Finite Scalar Quantization (FSQ), a simple per-dimension discretization with smooth optimization. At about 2.62 Hz (tokens/s), TASLA consistently improves prosody and achieves competitive quality over TASTE on in-domain (LibriSpeech) and OOD (EXPRESSO, Voxceleb) sets. We further demonstrate that dynamic layer mixing is correlated with spectral flux and explains why MLDA preserves prosody under a low frame rate with extreme feature compression.

cs.SD↗

arXiv2025

Emilia: A Large-Scale, Extensive, Multilingual, and Diverse Dataset for Speech Generation

Recent advancements in speech generation have been driven by large-scale training datasets. However, current models struggle to capture the spontaneity and variability inherent in real-world human speech, as they are primarily trained on audio-book datasets limited to formal, read-aloud speaking styles. To address this limitation, we introduce Emilia-Pipe, an open-source preprocessing pipeline designed to extract high-quality training data from valuable yet under-explored in-the-wild sources that capture spontaneous human speech in real-world contexts. Using Emilia-Pipe, we construct Emilia, which comprises over 101k hours of speech across six languages: English, Chinese, German, French, Japanese, and Korean. Furthermore, we expand Emilia to Emilia-Large, a dataset exceeding 216k hours, making it one of the largest open-source speech generation resources available. Extensive experiments show that Emilia-trained models produce markedly more spontaneous, human-like speech than those trained on traditional audio-book datasets, while matching their intelligibility. These models better capture diverse speaker timbres and the full spectrum of real-world conversational styles. Our work also highlights the importance of scaling dataset size for advancing speech generation performance and validates the effectiveness of Emilia for both multilingual and crosslingual speech generation tasks.

cs.SD↗

arXiv2025

DualCodec: A Low-Frame-Rate, Semantically-Enhanced Neural Audio Codec for Speech Generation

Neural audio codecs form the foundational building blocks for language model (LM)-based speech generation. Typically, there is a trade-off between frame rate and audio quality. This study introduces a low-frame-rate, semantically enhanced codec model. Existing approaches distill semantically rich self-supervised (SSL) representations into the first-layer codec tokens. This work proposes DualCodec, a dual-stream encoding approach that integrates SSL and waveform representations within an end-to-end codec framework. In this setting, DualCodec enhances the semantic information in the first-layer codec and enables the codec system to maintain high audio quality while operating at a low frame rate. Note that a low-frame-rate codec improves the efficiency of speech generation. Experimental results on audio codec and speech generation tasks confirm the effectiveness of the proposed DualCodec compared to state-of-the-art codec systems, such as Mimi Codec, SpeechTokenizer, DAC, and Encodec. Demos are available at: https://dualcodec.github.io, code is available at: https://github.com/jiaqili3/DualCodec

cs.SD↗

arXiv2025

SingVERSE: A Diverse, Real-World Benchmark for Singing Voice Enhancement

This paper presents a benchmark for singing voice enhancement. The development of singing voice enhancement is limited by the lack of realistic evaluation data. To address this gap, this paper introduces SingVERSE, the first real-world benchmark for singing voice enhancement, covering diverse acoustic scenarios and providing paired, studio-quality clean references. Leveraging SingVERSE, we conduct a comprehensive evaluation of state-of-the-art models and uncover a consistent trade-off between perceptual quality and intelligibility. Finally, we show that training on in-domain singing data substantially improves enhancement performance without degrading speech capabilities, establishing a simple yet effective path forward. This work offers the community a foundational benchmark together with critical insights to guide future advances in this underexplored domain. Demopage: https://singverse.github.io

cs.SD↗

arXiv2025

AnyAccomp: Generalizable Accompaniment Generation via Quantized Melodic Bottleneck

Singing Accompaniment Generation (SAG) is the process of generating instrumental music for a given clean vocal input. However, existing SAG techniques use source-separated vocals as input and overfit to separation artifacts. This creates a critical train-test mismatch, leading to failure on clean, real-world vocal inputs. We introduce AnyAccomp, a framework that resolves this by decoupling accompaniment generation from source-dependent artifacts. AnyAccomp first employs a quantized melodic bottleneck, using a chromagram and a VQ-VAE to extract a discrete and timbre-invariant representation of the core melody. A subsequent flow-matching model then generates the accompaniment conditioned on these robust codes. Experiments show AnyAccomp achieves competitive performance on separated-vocal benchmarks while significantly outperforming baselines on generalization test sets of clean studio vocals and, notably, solo instrumental tracks. This demonstrates a qualitative leap in generalization, enabling robust accompaniment for instruments - a task where existing models completely fail - and paving the way for more versatile music co-creation tools. Demo audio and code: https://anyaccomp.github.io

cs.SD↗

arXiv2025

Over-the-Air Adversarial Attack Detection: from Datasets to Defenses

Automatic Speaker Verification (ASV) systems can be used for voice-enabled applications for identity verification. However, recent studies have exposed these systems' vulnerabilities to both over-the-line (OTL) and over-the-air (OTA) adversarial attacks. Although various detection methods have been proposed to counter these threats, they have not been thoroughly tested due to the lack of a comprehensive data set. To address this gap, we developed the AdvSV 2.0 dataset, which contains 628k samples with a total duration of 800 hours. This dataset incorporates classical adversarial attack algorithms, ASV systems, and encompasses both OTL and OTA scenarios. Furthermore, we introduce a novel adversarial attack method based on a Neural Replay Simulator (NRS), which enhances the potency of adversarial OTA attacks, thereby presenting a greater threat to ASV systems. To defend against these attacks, we propose CODA-OCC, a contrastive learning approach within the one-class classification framework. Experimental results show that CODA-OCC achieves an EER of 11.2% and an AUC of 0.95 on the AdvSV 2.0 dataset, outperforming several state-of-the-art detection methods.

eess.AS↗

arXiv2025

Audio Deepfake Verification

With the rapid development of deepfake technology, simply making a binary judgment of true or false on audio is no longer sufficient to meet practical needs. Accurately determining the specific deepfake method has become crucial. This paper introduces the Audio Deepfake Verification (ADV) task, effectively addressing the limitations of existing deepfake source tracing methods in closed-set scenarios, aiming to achieve open-set deepfake source tracing. Meanwhile, the Audity dual-branch architecture is proposed, extracting deepfake features from two dimensions: audio structure and generation artifacts. Experimental results show that the dual-branch Audity architecture outperforms any single-branch configuration, and it can simultaneously achieve excellent performance in both deepfake detection and verification tasks.

eess.AS↗

arXiv2025

Noro: Noise-Robust One-shot Voice Conversion with Hidden Speaker Representation Learning

The effectiveness of one-shot voice conversion (VC) decreases in real-world scenarios where reference speeches, which are often sourced from the internet, contain various disturbances like background noise. To address this issue, we introduce Noro, a noise-robust one-shot VC system. Noro features innovative components tailored for VC using noisy reference speeches, including a dual-branch reference encoding module and a noise-agnostic contrastive speaker loss. Experimental results demonstrate that Noro outperforms our baseline system in both clean and noisy scenarios, highlighting its efficacy for real-world applications. Additionally, we investigate the hidden speaker representation capabilities of our baseline system by repurposing its reference encoder as a speaker encoder. The results show that it is competitive with several advanced self-supervised learning models for speaker representation under the SUPERB settings, highlighting the potential for advancing speaker representation learning through one-shot VC tasks.

cs.SD↗

arXiv2025

TaDiCodec: Text-aware Diffusion Speech Tokenizer for Speech Language Modeling

Speech tokenizers serve as foundational components for speech language models, yet current designs exhibit several limitations, including: 1) dependence on multi-layer residual vector quantization structures or high frame rates, 2) reliance on auxiliary pre-trained models for semantic distillation, and 3) requirements for complex two-stage training processes. In this work, we introduce the Text-aware Diffusion Transformer Speech Codec (TaDiCodec), a novel approach designed to overcome these challenges. TaDiCodec employs end-to-end optimization for quantization and reconstruction through a diffusion autoencoder, while integrating text guidance into the diffusion decoder to enhance reconstruction quality and achieve optimal compression. TaDiCodec achieves an extremely low frame rate of 6.25 Hz and a corresponding bitrate of 0.0875 kbps with a single-layer codebook for 24 kHz speech, while maintaining superior performance on critical speech generation evaluation metrics such as Word Error Rate (WER), speaker similarity (SIM), and speech quality (UTMOS). Notably, TaDiCodec employs a single-stage, end-to-end training paradigm, and obviating the need for auxiliary pre-trained models. We also validate the compatibility of TaDiCodec in language model based zero-shot text-to-speech with both autoregressive modeling and masked generative modeling, demonstrating its effectiveness and efficiency for speech language modeling, as well as a significantly small reconstruction-generation gap. We will open source our code and model checkpoints. Audio samples are are available at https:/tadicodec.github.io/. We release code and model checkpoints at https:/github.com/HeCheng0625/Diffusion-Speech-Tokenizer.

cs.SD↗

arXiv2025

Neurodyne: Neural Pitch Manipulation with Representation Learning and Cycle-Consistency GAN

Pitch manipulation is the process of producers adjusting the pitch of an audio segment to a specific key and intonation, which is essential in music production. Neural-network-based pitch-manipulation systems have been popular in recent years due to their superior synthesis quality compared to classical DSP methods. However, their performance is still limited due to their inaccurate feature disentanglement using source-filter models and the lack of paired in- and out-of-tune training data. This work proposes Neurodyne to address these issues. Specifically, Neurodyne uses adversarial representation learning to learn a pitch-independent latent representation to avoid inaccurate disentanglement and cycle-consistency training to create paired training data implicitly. Experimental results on global-key and template-based pitch manipulation demonstrate the effectiveness of the proposed system, marking improved synthesis quality while maintaining the original singer identity.

cs.SD↗

arXiv2025

NVSpeech: An Integrated and Scalable Pipeline for Human-Like Speech Modeling with Paralinguistic Vocalizations

Paralinguistic vocalizations-including non-verbal sounds like laughter and breathing, as well as lexicalized interjections such as "uhm" and "oh"-are integral to natural spoken communication. Despite their importance in conveying affect, intent, and interactional cues, such cues remain largely overlooked in conventional automatic speech recognition (ASR) and text-to-speech (TTS) systems. We present NVSpeech, an integrated and scalable pipeline that bridges the recognition and synthesis of paralinguistic vocalizations, encompassing dataset construction, ASR modeling, and controllable TTS. (1) We introduce a manually annotated dataset of 48,430 human-spoken utterances with 18 word-level paralinguistic categories. (2) We develop the paralinguistic-aware ASR model, which treats paralinguistic cues as inline decodable tokens (e.g., "You're so funny [Laughter]"), enabling joint lexical and non-verbal transcription. This model is then used to automatically annotate a large corpus, the first large-scale Chinese dataset of 174,179 utterances (573 hours) with word-level alignment and paralingustic cues. (3) We finetune zero-shot TTS models on both human- and auto-labeled data to enable explicit control over paralinguistic vocalizations, allowing context-aware insertion at arbitrary token positions for human-like speech synthesis. By unifying the recognition and generation of paralinguistic vocalizations, NVSpeech offers the first open, large-scale, word-level annotated pipeline for expressive speech modeling in Mandarin, integrating recognition and synthesis in a scalable and controllable manner. Dataset and audio demos are available at https://nvspeech170k.github.io/.

cs.SD↗

arXiv2025

Advancing Zero-shot Text-to-Speech Intelligibility across Diverse Domains via Preference Alignment

Modern zero-shot text-to-speech (TTS) systems, despite using extensive pre-training, often struggle in challenging scenarios such as tongue twisters, repeated words, code-switching, and cross-lingual synthesis, leading to intelligibility issues. To address these limitations, this paper leverages preference alignment techniques, which enable targeted construction of out-of-pretraining-distribution data to enhance performance. We introduce a new dataset, named the Intelligibility Preference Speech Dataset (INTP), and extend the Direct Preference Optimization (DPO) framework to accommodate diverse TTS architectures. After INTP alignment, in addition to intelligibility, we observe overall improvements including naturalness, similarity, and audio quality for multiple TTS models across diverse domains. Based on that, we also verify the weak-to-strong generalization ability of INTP for more intelligible models such as CosyVoice 2 and Ints. Moreover, we showcase the potential for further improvements through iterative alignment based on Ints. Audio samples are available at https://intalign.github.io/.

cs.SD↗

arXiv2025

Solid State Bus-Comp: A Large-Scale and Diverse Dataset for Dynamic Range Compressor Virtual Analog Modeling

Virtual Analog (VA) modeling aims to simulate the behavior of hardware circuits via algorithms to replicate their tone digitally. Dynamic Range Compressor (DRC) is an audio processing module that controls the dynamics of a track by reducing and amplifying the volumes of loud and quiet sounds, which is essential in music production. In recent years, neural-network-based VA modeling has shown great potential in producing high-fidelity models. However, due to the lack of data quantity and diversity, their generalization ability in different parameter settings and input sounds is still limited. To tackle this problem, we present Solid State Bus-Comp, the first large-scale and diverse dataset for modeling the classical VCA compressor -- SSL 500 G-Bus. Specifically, we manually collected 175 unmastered songs from the Cambridge Multitrack Library. We recorded the compressed audio in 220 parameter combinations, resulting in an extensive 2528-hour dataset with diverse genres, instruments, tempos, and keys. Moreover, to facilitate the use of our proposed dataset, we conducted benchmark experiments in various open-sourced black-box and grey-box models, as well as white-box plugins. We also conducted ablation studies in different data subsets to illustrate the effectiveness of the improved data diversity and quantity. The dataset and demos are on our project page: https://www.yichenggu.com/SolidStateBusComp/.

cs.SD↗

arXiv2025

SingNet: Towards a Large-Scale, Diverse, and In-the-Wild Singing Voice Dataset

The lack of a publicly-available large-scale and diverse dataset has long been a significant bottleneck for singing voice applications like Singing Voice Synthesis (SVS) and Singing Voice Conversion (SVC). To tackle this problem, we present SingNet, an extensive, diverse, and in-the-wild singing voice dataset. Specifically, we propose a data processing pipeline to extract ready-to-use training data from sample packs and songs on the internet, forming 3000 hours of singing voices in various languages and styles. Furthermore, to facilitate the use and demonstrate the effectiveness of SingNet, we pre-train and open-source various state-of-the-art (SOTA) models on Wav2vec2, BigVGAN, and NSF-HiFiGAN based on our collected singing voice data. We also conduct benchmark experiments on Automatic Lyric Transcription (ALT), Neural Vocoder, and Singing Voice Conversion (SVC). Audio demos are available at: https://singnet-dataset.github.io/.

cs.SD↗

arXiv2025

LOKI: A Comprehensive Synthetic Data Detection Benchmark using Large Multimodal Models

With the rapid development of AI-generated content, the future internet may be inundated with synthetic data, making the discrimination of authentic and credible multimodal data increasingly challenging. Synthetic data detection has thus garnered widespread attention, and the performance of large multimodal models (LMMs) in this task has attracted significant interest. LMMs can provide natural language explanations for their authenticity judgments, enhancing the explainability of synthetic content detection. Simultaneously, the task of distinguishing between real and synthetic data effectively tests the perception, knowledge, and reasoning capabilities of LMMs. In response, we introduce LOKI, a novel benchmark designed to evaluate the ability of LMMs to detect synthetic data across multiple modalities. LOKI encompasses video, image, 3D, text, and audio modalities, comprising 18K carefully curated questions across 26 subcategories with clear difficulty levels. The benchmark includes coarse-grained judgment and multiple-choice questions, as well as fine-grained anomaly selection and explanation tasks, allowing for a comprehensive analysis of LMMs. We evaluated 22 open-source LMMs and 6 closed-source models on LOKI, highlighting their potential as synthetic data detectors and also revealing some limitations in the development of LMM capabilities. More information about LOKI can be found at https://opendatalab.github.io/LOKI/

cs.CV↗

arXiv2025

Solla: Towards a Speech-Oriented LLM That Hears Acoustic Context

Large Language Models (LLMs) have recently shown remarkable ability to process not only text but also multimodal inputs such as speech and audio. However, most existing models primarily focus on analyzing input signals using text instructions, overlooking scenarios in which speech instructions and audio are mixed and serve as inputs to the model. To address these challenges, we introduce Solla, a novel framework designed to understand speech-based questions and hear the acoustic context concurrently. Solla incorporates an audio tagging module to effectively identify and represent audio events, as well as an ASR-assisted prediction method to improve comprehension of spoken content. To rigorously evaluate Solla and other publicly available models, we propose a new benchmark dataset called SA-Eval, which includes three tasks: audio event classification, audio captioning, and audio question answering. SA-Eval has diverse speech instruction with various speaking styles, encompassing two difficulty levels, easy and hard, to capture the range of real-world acoustic conditions. Experimental results show that Solla performs on par with or outperforms baseline models on both the easy and hard test sets, underscoring its effectiveness in jointly understanding speech and audio.

eess.AS↗

arXiv2025

Overview of the Amphion Toolkit (v0.2)

Amphion is an open-source toolkit for Audio, Music, and Speech Generation, designed to lower the entry barrier for junior researchers and engineers in these fields. It provides a versatile framework that supports a variety of generation tasks and models. In this report, we introduce Amphion v0.2, the second major release developed in 2024. This release features a 100K-hour open-source multilingual dataset, a robust data preparation pipeline, and novel models for tasks such as text-to-speech, audio coding, and voice conversion. Furthermore, the report includes multiple tutorials that guide users through the functionalities and usage of the newly released models.

cs.SD↗

arXiv2025

Vevo: Controllable Zero-Shot Voice Imitation with Self-Supervised Disentanglement

The imitation of voice, targeted on specific speech attributes such as timbre and speaking style, is crucial in speech generation. However, existing methods rely heavily on annotated data, and struggle with effectively disentangling timbre and style, leading to challenges in achieving controllable generation, especially in zero-shot scenarios. To address these issues, we propose Vevo, a versatile zero-shot voice imitation framework with controllable timbre and style. Vevo operates in two core stages: (1) Content-Style Modeling: Given either text or speech's content tokens as input, we utilize an autoregressive transformer to generate the content-style tokens, which is prompted by a style reference; (2) Acoustic Modeling: Given the content-style tokens as input, we employ a flow-matching transformer to produce acoustic representations, which is prompted by a timbre reference. To obtain the content and content-style tokens of speech, we design a fully self-supervised approach that progressively decouples the timbre, style, and linguistic content of speech. Specifically, we adopt VQ-VAE as the tokenizer for the continuous hidden features of HuBERT. We treat the vocabulary size of the VQ-VAE codebook as the information bottleneck, and adjust it carefully to obtain the disentangled speech representations. Solely self-supervised trained on 60K hours of audiobook speech data, without any fine-tuning on style-specific corpora, Vevo matches or surpasses existing methods in accent and emotion conversion tasks. Additionally, Vevo's effectiveness in zero-shot voice conversion and text-to-speech tasks further demonstrates its strong generalization and versatility. Audio samples are available at https://versavoice.github.io.

cs.SD↗