Yuancheng Wang

Yuancheng Wang, Haoyue Zhan, Liwei Liu et al.

The recent large-scale text-to-speech (TTS) systems are usually grouped as autoregressive and non-autoregressive systems. The autoregressive systems implicitly model duration but exhibit certain deficiencies in robustness and lack of duration controllability. Non-autoregressive systems require explicit alignment information between text and speech during training and predict durations for linguistic units (e.g. phone), which may compromise their naturalness. In this paper, we introduce Masked Generative Codec Transformer (MaskGCT), a fully non-autoregressive TTS model that eliminates the need for explicit alignment information between text and speech supervision, as well as phone-level duration prediction. MaskGCT is a two-stage model: in the first stage, the model uses text to predict semantic tokens extracted from a speech self-supervised learning (SSL) model, and in the second stage, the model predicts acoustic tokens conditioned on these semantic tokens. MaskGCT follows the mask-and-predict learning paradigm. During training, MaskGCT learns to predict masked semantic or acoustic tokens based on given conditions and prompts. During inference, the model generates tokens of a specified length in a parallel manner. Experiments with 100K hours of in-the-wild speech demonstrate that MaskGCT outperforms the current state-of-the-art zero-shot TTS systems in terms of quality, similarity, and intelligibility. Audio samples are available at https://maskgct.github.io/. We release our code and model checkpoints at https://github.com/open-mmlab/Amphion/blob/main/models/tts/maskgct.

Haorui He, Zengqiang Shang, Chaoren Wang et al.

Recent advancements in speech generation models have been significantly driven by the use of large-scale training data. However, producing highly spontaneous, human-like speech remains a challenge due to the scarcity of large, diverse, and spontaneous speech datasets. In response, we introduce Emilia, the first large-scale, multilingual, and diverse speech generation dataset. Emilia starts with over 101k hours of speech across six languages, covering a wide range of speaking styles to enable more natural and spontaneous speech generation. To facilitate the scale-up of Emilia, we also present Emilia-Pipe, the first open-source preprocessing pipeline designed to efficiently transform raw, in-the-wild speech data into high-quality training data with speech annotations. Experimental results demonstrate the effectiveness of both Emilia and Emilia-Pipe. Demos are available at: https://emilia-dataset.github.io/Emilia-Demo-Page/.

Yuancheng Wang, Zeqian Ju, Xu Tan et al.

Audio editing is applicable for various purposes, such as adding background sound effects, replacing a musical instrument, and repairing damaged audio. Recently, some diffusion-based methods achieved zero-shot audio editing by using a diffusion and denoising process conditioned on the text description of the output audio. However, these methods still have some problems: 1) they have not been trained on editing tasks and cannot ensure good editing effects; 2) they can erroneously modify audio segments that do not require editing; 3) they need a complete description of the output audio, which is not always available or necessary in practical scenarios. In this work, we propose AUDIT, an instruction-guided audio editing model based on latent diffusion models. Specifically, AUDIT has three main design features: 1) we construct triplet training data (instruction, input audio, output audio) for different audio editing tasks and train a diffusion model using instruction and input (to be edited) audio as conditions and generating output (edited) audio; 2) it can automatically learn to only modify segments that need to be edited by comparing the difference between the input and output audio; 3) it only needs edit instructions instead of full target audio descriptions as text input. AUDIT achieves state-of-the-art results in both objective and subjective metrics for several audio editing tasks (e.g., adding, dropping, replacement, inpainting, super-resolution). Demo samples are available at https://audit-demo.github.io/.

Junan Zhang, Jing Yang, Zihao Fang et al.

We introduce AnyEnhance, a unified generative model for voice enhancement that processes both speech and singing voices. Based on a masked generative model, AnyEnhance is capable of handling both speech and singing voices, supporting a wide range of enhancement tasks including denoising, dereverberation, declipping, super-resolution, and target speaker extraction, all simultaneously and without fine-tuning. AnyEnhance introduces a prompt-guidance mechanism for in-context learning, which allows the model to natively accept a reference speaker's timbre. In this way, it could boost enhancement performance when a reference audio is available and enable the target speaker extraction task without altering the underlying architecture. Moreover, we also introduce a self-critic mechanism into the generative process for masked generative models, yielding higher-quality outputs through iterative self-assessment and refinement. Extensive experiments on various enhancement tasks demonstrate AnyEnhance outperforms existing methods in terms of both objective metrics and subjective listening tests. Demo audios are publicly available at https://amphionspace.github.io/anyenhance. An open-source implementation is provided at https://github.com/viewfinder-annn/anyenhance-v1-ccf-aatc.

Haorui He, Zengqiang Shang, Chaoren Wang et al.

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.

Jiaqi Li, Xueyao Zhang, Yuancheng Wang et al.

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.

Xueyao Zhang, Chaoren Wang, Huan Liao et al.

Aligning large generative models with human feedback is a critical challenge. In speech synthesis, this is particularly pronounced due to the lack of a large-scale human preference dataset, which hinders the development of models that truly align with human perception. To address this, we introduce SpeechJudge, a comprehensive suite comprising a dataset, a benchmark, and a reward model centered on naturalness--one of the most fundamental subjective metrics for speech synthesis. First, we present SpeechJudge-Data, a large-scale human feedback corpus of 99K speech pairs. The dataset is constructed using a diverse set of advanced zero-shot text-to-speech (TTS) models across diverse speech styles and multiple languages, with human annotations for both intelligibility and naturalness preference. From this, we establish SpeechJudge-Eval, a challenging benchmark for speech naturalness judgment. Our evaluation reveals that existing metrics and AudioLLMs struggle with this task; the leading model, Gemini-2.5-Flash, achieves less than 70% agreement with human judgment, highlighting a significant gap for improvement. To bridge this gap, we develop SpeechJudge-GRM, a generative reward model (GRM) based on Qwen2.5-Omni-7B. It is trained on SpeechJudge-Data via a two-stage post-training process: Supervised Fine-Tuning (SFT) with Chain-of-Thought rationales followed by Reinforcement Learning (RL) with GRPO on challenging cases. On the SpeechJudge-Eval benchmark, the proposed SpeechJudge-GRM demonstrates superior performance, achieving 77.2% accuracy (and 79.4% after inference-time scaling @10) compared to a classic Bradley-Terry reward model (72.7%). Furthermore, SpeechJudge-GRM can be also employed as a reward function during the post-training of speech generation models to facilitate their alignment with human preferences.

Zeqian Ju, Yuancheng Wang, Kai Shen et al.

While recent large-scale text-to-speech (TTS) models have achieved significant progress, they still fall short in speech quality, similarity, and prosody. Considering speech intricately encompasses various attributes (e.g., content, prosody, timbre, and acoustic details) that pose significant challenges for generation, a natural idea is to factorize speech into individual subspaces representing different attributes and generate them individually. Motivated by it, we propose NaturalSpeech 3, a TTS system with novel factorized diffusion models to generate natural speech in a zero-shot way. Specifically, 1) we design a neural codec with factorized vector quantization (FVQ) to disentangle speech waveform into subspaces of content, prosody, timbre, and acoustic details; 2) we propose a factorized diffusion model to generate attributes in each subspace following its corresponding prompt. With this factorization design, NaturalSpeech 3 can effectively and efficiently model intricate speech with disentangled subspaces in a divide-and-conquer way. Experiments show that NaturalSpeech 3 outperforms the state-of-the-art TTS systems on quality, similarity, prosody, and intelligibility, and achieves on-par quality with human recordings. Furthermore, we achieve better performance by scaling to 1B parameters and 200K hours of training data.

Detai Xin, Xu Tan, Kai Shen et al.

We present RALL-E, a robust language modeling method for text-to-speech (TTS) synthesis. While previous work based on large language models (LLMs) shows impressive performance on zero-shot TTS, such methods often suffer from poor robustness, such as unstable prosody (weird pitch and rhythm/duration) and a high word error rate (WER), due to the autoregressive prediction style of language models. The core idea behind RALL-E is chain-of-thought (CoT) prompting, which decomposes the task into simpler steps to enhance the robustness of LLM-based TTS. To accomplish this idea, RALL-E first predicts prosody features (pitch and duration) of the input text and uses them as intermediate conditions to predict speech tokens in a CoT style. Second, RALL-E utilizes the predicted duration prompt to guide the computing of self-attention weights in Transformer to enforce the model to focus on the corresponding phonemes and prosody features when predicting speech tokens. Results of comprehensive objective and subjective evaluations demonstrate that, compared to a powerful baseline method VALL-E, RALL-E significantly improves the WER of zero-shot TTS from $5.6\%$ (without reranking) and $1.7\%$ (with reranking) to $2.5\%$ and $1.0\%$, respectively. Furthermore, we demonstrate that RALL-E correctly synthesizes sentences that are hard for VALL-E and reduces the error rate from $68\%$ to $4\%$.

Yuancheng Wang, Jiachen Zheng, Junan Zhang et al.

We introduce Metis, a foundation model for unified speech generation. Unlike previous task-specific or multi-task models, Metis follows a pre-training and fine-tuning paradigm. It is pre-trained on large-scale unlabeled speech data using masked generative modeling and then fine-tuned to adapt to diverse speech generation tasks. Specifically, 1) Metis utilizes two discrete speech representations: SSL tokens derived from speech self-supervised learning (SSL) features, and acoustic tokens directly quantized from waveforms. 2) Metis performs masked generative pre-training on SSL tokens, utilizing 300K hours of diverse speech data, without any additional condition. 3) Through fine-tuning with task-specific conditions, Metis achieves efficient adaptation to various speech generation tasks while supporting multimodal input, even when using limited data and trainable parameters. Experiments demonstrate that Metis can serve as a foundation model for unified speech generation: Metis outperforms state-of-the-art task-specific or multi-task systems across five speech generation tasks, including zero-shot text-to-speech, voice conversion, target speaker extraction, speech enhancement, and lip-to-speech, even with fewer than 20M trainable parameters or 300 times less training data. Audio samples are are available at https://metis-demo.github.io/.

Xueyao Zhang, Junan Zhang, Yuancheng Wang et al.

Controllable human voice generation, particularly for expressive domains like singing, remains a significant challenge. This paper introduces Vevo2, a unified framework for controllable speech and singing voice generation. To tackle issues like the scarcity of annotated singing data and to enable flexible controllability, Vevo2 introduces two audio tokenizers: (1) a music-notation-free prosody tokenizer that captures prosody and melody from speech, singing, and even instrumental sounds, and (2) a low-frame-rate (12.5 Hz) content-style tokenizer that encodes linguistic content, prosody, and style for both speech and singing, while enabling timbre disentanglement. Vevo2 consists of an auto-regressive (AR) content-style modeling stage, which aims to enable controllability over text, prosody, and style, as well as a flow-matching acoustic modeling stage that allows for timbre control. Particularly, during pre-training of the AR model, we propose both explicit and implicit prosody learning strategies to bridge speech and singing voice. Moreover, to further enhance the AR model's ability to follow text and prosody, we design a multi-objective post-training task that integrates both intelligibility and prosody similarity alignment. Experimental results show that the unified modeling in Vevo2 brings mutual benefits to both speech and singing voice generation. Additionally, Vevo2's effectiveness across a wide range of synthesis, conversion, and editing tasks for both speech and singing further demonstrates its strong generalization ability and versatility. Audio samples are are available at https://versasinger.github.io/.

Haorui He, Yuchen Song, Yuancheng Wang et al.

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.

Chuanfei Hu, Tianyi Xia, Ying Cui et al.

Multi-phase liver contrast-enhanced computed tomography (CECT) images convey the complementary multi-phase information for liver tumor segmentation (LiTS), which are crucial to assist the diagnosis of liver cancer clinically. However, the performances of existing multi-phase liver tumor segmentation (MPLiTS)-based methods suffer from redundancy and weak interpretability, % of the fused result, resulting in the implicit unreliability of clinical applications. In this paper, we propose a novel trustworthy multi-phase liver tumor segmentation (TMPLiTS), which is a unified framework jointly conducting segmentation and uncertainty estimation. The trustworthy results could assist the clinicians to make a reliable diagnosis. Specifically, Dempster-Shafer Evidence Theory (DST) is introduced to parameterize the segmentation and uncertainty as evidence following Dirichlet distribution. The reliability of segmentation results among multi-phase CECT images is quantified explicitly. Meanwhile, a multi-expert mixture scheme (MEMS) is proposed to fuse the multi-phase evidences, which can guarantee the effect of fusion procedure based on theoretical analysis. Experimental results demonstrate the superiority of TMPLiTS compared with the state-of-the-art methods. Meanwhile, the robustness of TMPLiTS is verified, where the reliable performance can be guaranteed against the perturbations.

Yuan Liu, Yuancheng Wang, Nan Li et al.

The main challenge of single image super resolution (SISR) is the recovery of high frequency details such as tiny textures. However, most of the state-of-the-art methods lack specific modules to identify high frequency areas, causing the output image to be blurred. We propose an attention-based approach to give a discrimination between texture areas and smooth areas. After the positions of high frequency details are located, high frequency compensation is carried out. This approach can incorporate with previously proposed SISR networks. By providing high frequency enhancement, better performance and visual effect are achieved. We also propose our own SISR network composed of DenseRes blocks. The block provides an effective way to combine the low level features and high level features. Extensive benchmark evaluation shows that our proposed method achieves significant improvement over the state-of-the-art works in SISR.

Dorian Cazau, Yuancheng Wang, Olivier Adam et al.

Many methods for automatic music transcription involves a multi-pitch estimation method that estimates an activity score for each pitch. A second processing step, called note segmentation, has to be performed for each pitch in order to identify the time intervals when the notes are played. In this study, a pitch-wise two-state on/off firstorder Hidden Markov Model (HMM) is developed for note segmentation. A complete parametrization of the HMM sigmoid function is proposed, based on its original regression formulation, including a parameter alpha of slope smoothing and beta? of thresholding contrast. A comparative evaluation of different note segmentation strategies was performed, differentiated according to whether they use a fixed threshold, called "Hard Thresholding" (HT), or a HMM-based thresholding method, called "Soft Thresholding" (ST). This evaluation was done following MIREX standards and using the MAPS dataset. Also, different transcription scenarios and recording natures were tested using three units of the Degradation toolbox. Results show that note segmentation through a HMM soft thresholding with a data-based optimization of the {alpha,beta} parameter couple significantly enhances transcription performance.