Yihua Wang

Xinhan Di, Zihao Chen, Yunming Liang et al.

Large-scale text-to-speech (TTS) models have made significant progress recently.However, they still fall short in the generation of Chinese dialectal speech. Toaddress this, we propose Bailing-TTS, a family of large-scale TTS models capable of generating high-quality Chinese dialectal speech. Bailing-TTS serves as a foundation model for Chinese dialectal speech generation. First, continual semi-supervised learning is proposed to facilitate the alignment of text tokens and speech tokens. Second, the Chinese dialectal representation learning is developed using a specific transformer architecture and multi-stage training processes. With the proposed design of novel network architecture and corresponding strategy, Bailing-TTS is able to generate Chinese dialectal speech from text effectively and efficiently. Experiments demonstrate that Bailing-TTS generates Chinese dialectal speech towards human-like spontaneous representation. Readers are encouraged to listen to demos at \url{https://c9412600.github.io/bltts_tech_report/index.html}.

Zihao Chen, Haomin Zhang, Xinhan Di et al.

Generating sound effects for product-level videos, where only a small amount of labeled data is available for diverse scenes, requires the production of high-quality sounds in few-shot settings. To tackle the challenge of limited labeled data in real-world scenes, we introduce YingSound, a foundation model designed for video-guided sound generation that supports high-quality audio generation in few-shot settings. Specifically, YingSound consists of two major modules. The first module uses a conditional flow matching transformer to achieve effective semantic alignment in sound generation across audio and visual modalities. This module aims to build a learnable audio-visual aggregator (AVA) that integrates high-resolution visual features with corresponding audio features at multiple stages. The second module is developed with a proposed multi-modal visual-audio chain-of-thought (CoT) approach to generate finer sound effects in few-shot settings. Finally, an industry-standard video-to-audio (V2A) dataset that encompasses various real-world scenarios is presented. We show that YingSound effectively generates high-quality synchronized sounds across diverse conditional inputs through automated evaluations and human studies. Project Page: \url{https://giantailab.github.io/yingsound/}

Ziqi Chen, Gongyu Chen, Yihua Wang et al.

Dialect speech embodies rich cultural and linguistic diversity, yet building text-to-speech (TTS) systems for dialects remains challenging due to scarce data, inconsistent orthographies, and complex phonetic variation. To address these issues, we present DiaMoE-TTS, a unified IPA-based framework that standardizes phonetic representations and resolves grapheme-to-phoneme ambiguities. Built upon the F5-TTS architecture, the system introduces a dialect-aware Mixture-of-Experts (MoE) to model phonological differences and employs parameter-efficient adaptation with Low-Rank Adaptors (LoRA) and Conditioning Adapters for rapid transfer to new dialects. Unlike approaches dependent on large-scale or proprietary resources, DiaMoE-TTS enables scalable, open-data-driven synthesis. Experiments demonstrate natural and expressive speech generation, achieving zero-shot performance on unseen dialects and specialized domains such as Peking Opera with only a few hours of data.

Yihua Wang, Qi Jia, Cong Xu et al.

Multimodal sarcasm detection is a complex task that requires distinguishing subtle complementary signals across modalities while filtering out irrelevant information. Many advanced methods rely on learning shortcuts from datasets rather than extracting intended sarcasm-related features. However, our experiments show that shortcut learning impairs the model's generalization in real-world scenarios. Furthermore, we reveal the weaknesses of current modality fusion strategies for multimodal sarcasm detection through systematic experiments, highlighting the necessity of focusing on effective modality fusion for complex emotion recognition. To address these challenges, we construct MUStARD++$^{R}$ by removing shortcut signals from MUStARD++. Then, a Multimodal Conditional Information Bottleneck (MCIB) model is introduced to enable efficient multimodal fusion for sarcasm detection. Experimental results show that the MCIB achieves the best performance without relying on shortcut learning.

Daniele Gammelli, Yihua Wang, Dennis Prak et al.

Bike-sharing systems are a rapidly developing mode of transportation and provide an efficient alternative to passive, motorized personal mobility. The asymmetric nature of bike demand causes the need for rebalancing bike stations, which is typically done during night time. To determine the optimal starting inventory level of a station for a given day, a User Dissatisfaction Function (UDF) models user pickups and returns as non-homogeneous Poisson processes with piece-wise linear rates. In this paper, we devise a deep generative model directly applicable in the UDF by introducing a variational Poisson recurrent neural network model (VP-RNN) to forecast future pickup and return rates. We empirically evaluate our approach against both traditional and learning-based forecasting methods on real trip travel data from the city of New York, USA, and show how our model outperforms benchmarks in terms of system efficiency and demand satisfaction. By explicitly focusing on the combination of decision-making algorithms with learning-based forecasting methods, we highlight a number of shortcomings in literature. Crucially, we show how more accurate predictions do not necessarily translate into better inventory decisions. By providing insights into the interplay between forecasts, model assumptions, and decisions, we point out that forecasts and decision models should be carefully evaluated and harmonized to optimally control shared mobility systems.