Search

Recent AIGC systems possess the capability to generate digital multimedia content based on human language instructions, such as text, image and video. However, when it comes to speech, existing methods related to human instruction-to-speech generation exhibit two limitations. Firstly, they require the division of inputs into content prompt (transcript) and description prompt (style and speaker), instead of directly supporting human instruction. This division is less natural in form and does not align with other AIGC models. Secondly, the practice of utilizing an independent description prompt to model speech style, without considering the transcript content, restricts the ability to control speech at a fine-grained level. To address these limitations, we propose VoxInstruct, a novel unified multilingual codec language modeling framework that extends traditional text-to-speech tasks into a general human instruction-to-speech task. Our approach enhances the expressiveness of human instruction-guided speech generation and aligns the speech generation paradigm with other modalities. To enable the model to automatically extract the content of synthesized speech from raw text instructions, we introduce speech semantic tokens as an intermediate representation for instruction-to-content guidance. We also incorporate multiple Classifier-Free Guidance (CFG) strategies into our codec language model, which strengthens the generated speech following human instructions. Furthermore, our model architecture and training strategies allow for the simultaneous support of combining speech prompt and descriptive human instruction for expressive speech synthesis, which is a first-of-its-kind attempt. Codes, models and demos are at: https://github.com/thuhcsi/VoxInstruct., Comment: Accepted by ACM Multimedia 2024

In recent years, foundation models (FMs) such as large language models (LLMs) and latent diffusion models (LDMs) have profoundly impacted diverse sectors, including music. This comprehensive review examines state-of-the-art (SOTA) pre-trained models and foundation models in music, spanning from representation learning, generative learning and multimodal learning. We first contextualise the significance of music in various industries and trace the evolution of AI in music. By delineating the modalities targeted by foundation models, we discover many of the music representations are underexplored in FM development. Then, emphasis is placed on the lack of versatility of previous methods on diverse music applications, along with the potential of FMs in music understanding, generation and medical application. By comprehensively exploring the details of the model pre-training paradigm, architectural choices, tokenisation, finetuning methodologies and controllability, we emphasise the important topics that should have been well explored, like instruction tuning and in-context learning, scaling law and emergent ability, as well as long-sequence modelling etc. A dedicated section presents insights into music agents, accompanied by a thorough analysis of datasets and evaluations essential for pre-training and downstream tasks. Finally, by underscoring the vital importance of ethical considerations, we advocate that following research on FM for music should focus more on such issues as interpretability, transparency, human responsibility, and copyright issues. The paper offers insights into future challenges and trends on FMs for music, aiming to shape the trajectory of human-AI collaboration in the music realm.

This paper presents the multi-speaker multi-lingual few-shot voice cloning system developed by THU-HCSI team for LIMMITS'24 Challenge. To achieve high speaker similarity and naturalness in both mono-lingual and cross-lingual scenarios, we build the system upon YourTTS and add several enhancements. For further improving speaker similarity and speech quality, we introduce speaker-aware text encoder and flow-based decoder with Transformer blocks. In addition, we denoise the few-shot data, mix up them with pre-training data, and adopt a speaker-balanced sampling strategy to guarantee effective fine-tuning for target speakers. The official evaluations in track 1 show that our system achieves the best speaker similarity MOS of 4.25 and obtains considerable naturalness MOS of 3.97., Comment: Accepted in Grand Challenge of ICASSP 2024

Graph neural networks (GNNs) have exhibited impressive performance in modeling graph data as exemplified in various applications. Recently, the GNN calibration problem has attracted increasing attention, especially in cost-sensitive scenarios. Previous work has gained empirical insights on the issue, and devised effective approaches for it, but theoretical supports still fall short. In this work, we shed light on the relationship between GNN calibration and nodewise similarity via theoretical analysis. A novel calibration framework, named SimCalib, is accordingly proposed to consider similarity between nodes at global and local levels. At the global level, the Mahalanobis distance between the current node and class prototypes is integrated to implicitly consider similarity between the current node and all nodes in the same class. At the local level, the similarity of node representation movement dynamics, quantified by nodewise homophily and relative degree, is considered. Informed about the application of nodewise movement patterns in analyzing nodewise behavior on the over-smoothing problem, we empirically present a possible relationship between over-smoothing and GNN calibration problem. Experimentally, we discover a correlation between nodewise similarity and model calibration improvement, in alignment with our theoretical results. Additionally, we conduct extensive experiments investigating different design factors and demonstrate the effectiveness of our proposed SimCalib framework for GNN calibration by achieving state-of-the-art performance on 14 out of 16 benchmarks.

Speech-driven 3D facial animation aims at generating facial movements that are synchronized with the driving speech, which has been widely explored recently. Existing works mostly neglect the person-specific talking style in generation, including facial expression and head pose styles. Several works intend to capture the personalities by fine-tuning modules. However, limited training data leads to the lack of vividness. In this work, we propose AdaMesh, a novel adaptive speech-driven facial animation approach, which learns the personalized talking style from a reference video of about 10 seconds and generates vivid facial expressions and head poses. Specifically, we propose mixture-of-low-rank adaptation (MoLoRA) to fine-tune the expression adapter, which efficiently captures the facial expression style. For the personalized pose style, we propose a pose adapter by building a discrete pose prior and retrieving the appropriate style embedding with a semantic-aware pose style matrix without fine-tuning. Extensive experimental results show that our approach outperforms state-of-the-art methods, preserves the talking style in the reference video, and generates vivid facial animation. The supplementary video and code will be available at https://adamesh.github.io., Comment: Project Page: https://adamesh.github.io

Zero-shot text-to-speech (TTS) synthesis aims to clone any unseen speaker's voice without adaptation parameters. By quantizing speech waveform into discrete acoustic tokens and modeling these tokens with the language model, recent language model-based TTS models show zero-shot speaker adaptation capabilities with only a 3-second acoustic prompt of an unseen speaker. However, they are limited by the length of the acoustic prompt, which makes it difficult to clone personal speaking style. In this paper, we propose a novel zero-shot TTS model with the multi-scale acoustic prompts based on a neural codec language model VALL-E. A speaker-aware text encoder is proposed to learn the personal speaking style at the phoneme-level from the style prompt consisting of multiple sentences. Following that, a VALL-E based acoustic decoder is utilized to model the timbre from the timbre prompt at the frame-level and generate speech. The experimental results show that our proposed method outperforms baselines in terms of naturalness and speaker similarity, and can achieve better performance by scaling out to a longer style prompt., Comment: Accepted bt ICASSP 2024

This paper presents an end-to-end high-quality singing voice synthesis (SVS) system that uses bidirectional encoder representation from Transformers (BERT) derived semantic embeddings to improve the expressiveness of the synthesized singing voice. Based on the main architecture of recently proposed VISinger, we put forward several specific designs for expressive singing voice synthesis. First, different from the previous SVS models, we use text representation of lyrics extracted from pre-trained BERT as additional input to the model. The representation contains information about semantics of the lyrics, which could help SVS system produce more expressive and natural voice. Second, we further introduce an energy predictor to stabilize the synthesized voice and model the wider range of energy variations that also contribute to the expressiveness of singing voice. Last but not the least, to attenuate the off-key issues, the pitch predictor is re-designed to predict the real to note pitch ratio. Both objective and subjective experimental results indicate that the proposed SVS system can produce singing voice with higher-quality outperforming VISinger.

The spontaneous behavior that often occurs in conversations makes speech more human-like compared to reading-style. However, synthesizing spontaneous-style speech is challenging due to the lack of high-quality spontaneous datasets and the high cost of labeling spontaneous behavior. In this paper, we propose a semi-supervised pre-training method to increase the amount of spontaneous-style speech and spontaneous behavioral labels. In the process of semi-supervised learning, both text and speech information are considered for detecting spontaneous behaviors labels in speech. Moreover, a linguistic-aware encoder is used to model the relationship between each sentence in the conversation. Experimental results indicate that our proposed method achieves superior expressive speech synthesis performance with the ability to model spontaneous behavior in spontaneous-style speech and predict reasonable spontaneous behavior from text., Comment: Accepted by INTERSPEECH 2023

Expressive speech synthesis is crucial for many human-computer interaction scenarios, such as audiobooks, podcasts, and voice assistants. Previous works focus on predicting the style embeddings at one single scale from the information within the current sentence. Whereas, context information in neighboring sentences and multi-scale nature of style in human speech are neglected, making it challenging to convert multi-sentence text into natural and expressive speech. In this paper, we propose MSStyleTTS, a style modeling method for expressive speech synthesis, to capture and predict styles at different levels from a wider range of context rather than a sentence. Two sub-modules, including multi-scale style extractor and multi-scale style predictor, are trained together with a FastSpeech 2 based acoustic model. The predictor is designed to explore the hierarchical context information by considering structural relationships in context and predict style embeddings at global-level, sentence-level and subword-level. The extractor extracts multi-scale style embedding from the ground-truth speech and explicitly guides the style prediction. Evaluations on both in-domain and out-of-domain audiobook datasets demonstrate that the proposed method significantly outperforms the three baselines. In addition, we conduct the analysis of the context information and multi-scale style representations that have never been discussed before., Comment: Accepted by IEEE/ACM Transactions on Audio, Speech, and Language Processing

Music-driven 3D dance generation has become an intensive research topic in recent years with great potential for real-world applications. Most existing methods lack the consideration of genre, which results in genre inconsistency in the generated dance movements. In addition, the correlation between the dance genre and the music has not been investigated. To address these issues, we propose a genre-consistent dance generation framework, GTN-Bailando. First, we propose the Genre Token Network (GTN), which infers the genre from music to enhance the genre consistency of long-term dance generation. Second, to improve the generalization capability of the model, the strategy of pre-training and fine-tuning is adopted.Experimental results on the AIST++ dataset show that the proposed dance generation framework outperforms state-of-the-art methods in terms of motion quality and genre consistency., Comment: Accepted by ICASSP2023.Demo page: https://im1eon.github.io/ICASSP23-GTNB-DG/

Recent advances in text-to-speech have significantly improved the expressiveness of synthesized speech. However, it is still challenging to generate speech with contextually appropriate and coherent speaking style for multi-sentence text in audiobooks. In this paper, we propose a context-aware coherent speaking style prediction method for audiobook speech synthesis. To predict the style embedding of the current utterance, a hierarchical transformer-based context-aware style predictor with a mixture attention mask is designed, considering both text-side context information and speech-side style information of previous speeches. Based on this, we can generate long-form speech with coherent style and prosody sentence by sentence. Objective and subjective evaluations on a Mandarin audiobook dataset demonstrate that our proposed model can generate speech with more expressive and coherent speaking style than baselines, for both single-sentence and multi-sentence test., Comment: Accepted by ICASSP 2023

Previous works on expressive speech synthesis focus on modelling the mono-scale style embedding from the current sentence or context, but the multi-scale nature of speaking style in human speech is neglected. In this paper, we propose a multi-scale speaking style modelling method to capture and predict multi-scale speaking style for improving the naturalness and expressiveness of synthetic speech. A multi-scale extractor is proposed to extract speaking style embeddings at three different levels from the ground-truth speech, and explicitly guide the training of a multi-scale style predictor based on hierarchical context information. Both objective and subjective evaluations on a Mandarin audiobooks dataset demonstrate that our proposed method can significantly improve the naturalness and expressiveness of the synthesized speech., Comment: Accepted by INTERSPEECH 2022

Previous works on expressive speech synthesis mainly focus on current sentence. The context in adjacent sentences is neglected, resulting in inflexible speaking style for the same text, which lacks speech variations. In this paper, we propose a hierarchical framework to model speaking style from context. A hierarchical context encoder is proposed to explore a wider range of contextual information considering structural relationship in context, including inter-phrase and inter-sentence relations. Moreover, to encourage this encoder to learn style representation better, we introduce a novel training strategy with knowledge distillation, which provides the target for encoder training. Both objective and subjective evaluations on a Mandarin lecture dataset demonstrate that the proposed method can significantly improve the naturalness and expressiveness of the synthesized speech., Comment: Accepted by ICASSP 2022

Bitcoin, one of the major cryptocurrencies, presents great opportunities and challenges with its tremendous potential returns accompanying high risks. The high volatility of Bitcoin and the complex factors affecting them make the study of effective price forecasting methods of great practical importance to financial investors and researchers worldwide. In this paper, we propose a novel approach called MRC-LSTM, which combines a Multi-scale Residual Convolutional neural network (MRC) and a Long Short-Term Memory (LSTM) to implement Bitcoin closing price prediction. Specifically, the Multi-scale residual module is based on one-dimensional convolution, which is not only capable of adaptive detecting features of different time scales in multivariate time series, but also enables the fusion of these features. LSTM has the ability to learn long-term dependencies in series, which is widely used in financial time series forecasting. By mixing these two methods, the model is able to obtain highly expressive features and efficiently learn trends and interactions of multivariate time series. In the study, the impact of external factors such as macroeconomic variables and investor attention on the Bitcoin price is considered in addition to the trading information of the Bitcoin market. We performed experiments to predict the daily closing price of Bitcoin (USD), and the experimental results show that MRC-LSTM significantly outperforms a variety of other network structures. Furthermore, we conduct additional experiments on two other cryptocurrencies, Ethereum and Litecoin, to further confirm the effectiveness of the MRC-LSTM in short-term forecasting for multivariate time series of cryptocurrencies.