Event extraction (EE) is a critical direction in the field of information extraction, laying an important foundation for the construction of structured knowledge bases. EE from text has received ample research and attention for years, yet there can be numerous real-world applications that require direct information acquisition from speech signals, online meeting minutes, interview summaries, press releases, etc. While EE from speech has remained under-explored, this paper fills the gap by pioneering a SpeechEE, defined as detecting the event predicates and arguments from a given audio speech. To benchmark the SpeechEE task, we first construct a large-scale high-quality dataset. Based on textual EE datasets under the sentence, document, and dialogue scenarios, we convert texts into speeches through both manual real-person narration and automatic synthesis, empowering the data with diverse scenarios, languages, domains, ambiences, and speaker styles. Further, to effectively address the key challenges in the task, we tailor an E2E SpeechEE system based on the encoder-decoder architecture, where a novel Shrinking Unit module and a retrieval-aided decoding mechanism are devised. Extensive experimental results on all SpeechEE subsets demonstrate the efficacy of the proposed model, offering a strong baseline for the task. At last, being the first work on this topic, we shed light on key directions for future research. Our codes and the benchmark datasets are open at https://SpeechEE.github.io/
论文检索
输入标题、作者或关键词，从 2,480 篇学术成果中精准定位
Talking Face Generation (TFG) reconstructs facial motions concerning lips given speech input, which aims to generate highquality, synchronized, and lip-readable videos. Previous efforts have achieved success in generating quality and synchronization, and recently, there has been an increasing focus on the importance of intelligibility. Despite these efforts, there remains a challenge in achieving a balance among quality, synchronization, and intelligibility, often resulting in trade-offs that compromise one aspect in favor of another. In light of this, we propose SyncTalklip, a novel dual-tower framework designed to overcome the challenges of synchronization while improving lip-reading performance. To enhance the performance of SyncTalklip in both synchronization and intelligibility, we design AV-SyncNet, a pre-trained multi-task model, aiming to achieve a dual-focus on synchronization and intelligibility. Moreover, we propose a novel cross-modal contrastive learning bringing audio and video closer to enhance synchronization. Experimental results demonstrate that SyncTalklip achieves state-of-the-art performance in quality, intelligibility, and synchronization. Furthermore, extensive experiments have demonstrated our model's generalizability across domains. The code and demo is available at https://sync-talklip.github.io.
3D facial animation has attracted considerable attention due to its extensive applications in the multimedia field. Audio-driven 3D facial animation has been widely explored with promising results. However, multi-modal 3D facial animation, especially text-guided 3D facial animation is rarely explored due to the lack of multi-modal 3D facial animation dataset. To fill this gap, we first construct a large-scale multi-modal 3D facial animation dataset, MMHead, which consists of 49 hours of 3D facial motion sequences, speech audios, and rich hierarchical text annotations. Each text annotation contains abstract action and emotion descriptions, fine-grained facial and head movements (i.e., expression and head pose) descriptions, and three possible scenarios that may cause such emotion. Concretely, we integrate five public 2D portrait video datasets, and propose an automatic pipeline to 1) reconstruct 3D facial motion sequences from monocular videos; and 2) obtain hierarchical text annotations with the help of AU detection and ChatGPT. Based on the MMHead dataset, we establish benchmarks for two new tasks: text-induced 3D talking head animation and text-to-3D facial motion generation. Moreover, a simple but efficient VQ-VAE-based method named MM2Face is proposed to unify the multi-modal information and generate diverse and plausible 3D facial motions, which achieves competitive results on both benchmarks. Extensive experiments and comprehensive analysis demonstrate the significant potential of our dataset and benchmarks in promoting the development of multi-modal 3D facial animation. The dataset will be released at: https://wsj-sjtu.github.io/MMHead/.
Dataset distillation, also known as dataset condensation, offers a possibility for compressing a large-scale dataset into a small-scale one (i.e., distilled dataset) while achieving similar performance during model training. This method effectively tackles the challenges of training efficiency and storage cost posed by the large-scale dataset. Existing dataset distillation methods can be categorized into Optimization-Oriented (OO)-based and Distribution-Matching (DM)-based methods. Since OO-based methods require bi-level optimization to alternately optimize the model and the distilled data, they face challenges due to high computational overhead in practical applications. Thus, DM-based methods have emerged as an alternative by aligning the prototypes of the distilled data to those of the original data. Although efficient, these methods overlook the diversity of the distilled data, which will limit the performance of evaluation tasks. In this paper, we propose a novel Diversified Semantic Distribution Matching (DSDM) approach for dataset distillation. To accurately capture semantic features, we first pre-train models for dataset distillation. Subsequently, we estimate the distribution of each category by calculating its prototype and covariance matrix, where the covariance matrix indicates the direction of semantic feature transformations for each category. Then, in addition to the prototypes, the covariance matrices are also matched to obtain more diversity for the distilled data. However, since the distilled data are optimized by multiple pre-trained models, the training process will fluctuate severely. Therefore, we match the distilled data of the current pre-trained model with the historical integrated prototypes. Experimental results demonstrate that our DSDM achieves state-of-the-art results on both image and speech datasets. Code is available at https://github.com/Li-Hongcheng/DSDM.
Movie Dubbing aims to convert scripts into speeches that align with the given movie clip in both temporal and emotional aspects while preserving the vocal timbre of one brief reference audio. The wide variations in emotion, pace, and environment that dubbed speech must exhibit to achieve real alignment make dubbing a complex task. Considering the limited scale of the movie dubbing datasets (due to copyright) and the interference from background noise, directly learning from movie dubbing datasets limits the pronunciation quality of learned models. To address this problem, we propose a two-stage dubbing method that allows the model to first learn pronunciation knowledge before practicing it in movie dubbing. In the first stage, we introduce a multi-task approach to pre-train a phoneme encoder on a large-scale text-speech corpus for learning clear and natural phoneme pronunciations. For the second stage, we devise a prosody consistency learning module to bridge the emotional expression with the phoneme-level dubbing prosody attributes (pitch and energy). Finally, we design a duration consistency reasoning module to align the dubbing duration with the lip movement. Extensive experiments demonstrate that our method outperforms several state-of-the-art methods on two primary benchmarks. The demos are available at https://speaker2dubber.github.io/.
Understanding the speaking style, such as the emotion of the interlocutor's speech, and responding with speech in an appropriate style is a natural occurrence in human conversations. However, technically, existing research on speech synthesis and speaking style captioning typically proceeds independently. In this work, an innovative framework, referred to as UniStyle, is proposed to incorporate both the capabilities of speaking style captioning and style-controllable speech synthesizing. Specifically, UniStyle consists of a UniConnector and a style prompt-based speech generator. The role of the UniConnector is to bridge the gap between different modalities, namely speech audio and text descriptions. It enables the generation of text descriptions with speech as input and the creation of style representations from text descriptions for speech synthesis with the speech generator. Besides, to overcome the issue of data scarcity, we propose a two-stage and semi-supervised training strategy, which reduces data requirements while boosting performance. Extensive experiments conducted on open-source corpora demonstrate that UniStyle achieves state-of-the-art performance in speaking style captioning and synthesizes expressive speech with various speaker timbres and speaking styles in a zero-shot manner.
MultiHateClip: A Multilingual Benchmark Dataset for Hateful Video Detection on YouTube and Bilibili
PDF ↗Hate speech is a pressing issue in modern society, with significant effects both online and offline. Recent research in hate speech detection has primarily centered on text-based media, largely overlooking multimodal content such as videos. Existing studies on hateful video datasets have predominantly focused on English content within a Western context and have been limited to binary labels (hateful or non-hateful), lacking detailed contextual information. This study presents MultiHateClip, an novel multilingual dataset created through hate lexicons and human annotation. It aims to enhance the detection of hateful videos on platforms such as YouTube and Bilibili, including content in both English and Chinese languages. Comprising 2,000 videos annotated for hatefulness, offensiveness, and normalcy, this dataset provides a cross-cultural perspective on gender-based hate speech. Through a detailed examination of human annotation results, we discuss the differences between Chinese and English hateful videos and underscore the importance of different modalities in hateful and offensive video analysis. Evaluations of state-of-the-art video classification models, such as VLM, GPT-4V and Qwen-VL, on MultiHateClip highlight the existing challenges in accurately distinguishing between hateful and offensive content and the urgent need for models that are both multimodally and culturally nuanced. MultiHateClip represents a foundational advance in enhancing hateful video detection by underscoring the necessity of a multimodal and culturally sensitive approach in combating online hate speech. Disclaimer: This paper contains sensitive content that may be disturbing to some readers.
Some video traffic carries harmful content, such as hate speech and child abuse, primarily encrypted and transmitted through Dynamic Adaptive Streaming over HTTP (DASH). Promptly identifying and intercepting traffic of harmful videos is crucial in network regulation. However, QUIC is becoming another DASH transport protocol in addition to TCP. On the other hand, complex network environments and diverse playback modes lead to significant distortions in traffic. The issues above have not been effectively addressed. This paper proposes a real-time identification method for DASH encrypted video traffic with distortion, named Zenith. We extract stable video segment sequences under various itags as video fingerprints to tackle resolution changes and propose a method of traffic fingerprint extraction under QUIC and VPN. Subsequently, simulating the sequence matching problem as a natural language problem, we propose Traffic Language Model (TLM), which can effectively address video data loss and retransmission. Finally, we propose a frequency dictionary to accelerate Zenith's speed further. Zenith significantly improves accuracy and speed compared to other SOTA methods in various complex scenarios, especially in QUIC, VPN, automatic resolution, and low bandwidth. Zenith requires traffic for just half a minute of video content to achieve precise identification, demonstrating its real-time effectiveness.
Speech-driven gesture generation aims at synthesizing a gesture sequence synchronized with the input speech signal. Previous methods leverage neural networks to directly map a compact audio representation to the gesture sequence, ignoring the semantic association of different modalities and failing to deal with salient gestures. In this paper, we propose a novel speech-driven gesture generation method by emphasizing the semantic consistency of salient posture. Specifically, we first learn a joint manifold space for the individual representation of audio and body pose to exploit the inherent semantic association between two modalities, and propose to enforce semantic consistency via a consistency loss. Furthermore, we emphasize the semantic consistency of salient postures by introducing a weakly-supervised detector to identify salient postures, and reweighting the consistency loss to focus more on learning the correspondence between salient postures and the high-level semantics of speech content. In addition, we propose to extract audio features dedicated to facial expression and body gesture separately, and design separate branches for face and body gesture synthesis. Extensive experimental results demonstrate the superiority of our method over the state-of-the-art approaches.
Recent progress in large-scale zero-shot speech synthesis has been significantly advanced by language models and diffusion models. However, the generation process of both methods is slow and computationally intensive. Efficient speech synthesis using a lower computing budget to achieve quality on par with previous work remains a significant challenge. In this paper, we present FlashSpeech, a large-scale zero-shot speech synthesis system with approximately 5% of the inference time compared with previous work. FlashSpeech is built on the latent consistency model and applies a novel adversarial consistency training approach that can train from scratch without the need for a pre-trained diffusion model as the teacher. Furthermore, a new prosody generator module enhances the diversity of prosody, making the rhythm of the speech sound more natural. The generation processes of FlashSpeech can be achieved efficiently with one or two sampling steps while maintaining high audio quality and high similarity to the audio prompt for zero-shot speech generation. Our experimental results demonstrate the superior performance of FlashSpeech. Notably, FlashSpeech can be about 20 times faster than other zero-shot speech synthesis systems while maintaining comparable performance in terms of voice quality and similarity. Furthermore, FlashSpeech demonstrates its versatility by efficiently performing tasks like voice conversion, speech editing, and diverse speech sampling. Audio samples can be found in https://flashspeech.github.io/
Voice is one of the most widely used media for information transmission in human society. While high-quality synthetic voices are extensively utilized in various applications, they pose significant risks to content security and trust building. Numerous studies have concentrated on AI-synthesized voice detection to mitigate these risks, with many claiming to achieve promising performance. However, recent research has demonstrated that fake voice detectors suffer from serious overfitting to speaker-irrelative features (SiFs) and cannot be used in real-world scenarios. In this paper, we analyze the limitations of existing fake voice detectors and propose a new design philosophy, guiding the detection model to prioritize learning human voice features rather than the difference between the human voice and the synthetic voice. Based on this philosophy, we propose a novel AI-synthesized voice detection framework named SiFSafer, which uses pre-trained speech representation models to enhance the learning of feature distribution in human voices and the adapter fine-tuning to optimize the performance. The evaluation shows that the average EERs of existing fake voice detectors in the ASVspoof datasets can exceed 20% if the SiFs like silence segments are removed, while SiFSafer achieves an EER of less than 8%, indicating that SiFSafer is robust to SiFs and strongly resistant to existing attacks.
Current co-speech motion generation approaches usually focus on upper body gestures following speech contents only, while lacking supporting the elaborate control of synergistic full-body motion based on text prompts, such as talking while walking. The major challenges lie in 1) the existing speech-to-motion datasets only involve highly limited full-body motions, making a wide range of common human activities out of training distribution; 2) these datasets also lack annotated user prompts. To address these challenges, we propose SynTalker, which utilizes the off-the-shelf text-to-motion dataset as an auxiliary for supplementing the missing full-body motion and prompts. The core technical contributions are two-fold. One is the multi-stage training process which obtains an aligned embedding space of motion, speech, and prompts despite the significant distributional mismatch in motion between speech-to-motion and text-to-motion datasets. Another is the diffusion-based conditional inference process, which utilizes the separate-then-combine strategy to realize fine-grained control of local body parts. Extensive experiments are conducted to verify that our approach supports precise and flexible control of synergistic full-body motion generation based on both speeches and user prompts, which is beyond the ability of existing approaches. Our code, pre-trained models, and videos are available at https://bohongchen.github.io/SynTalker-Page/.
Generative AI technologies, including text-to-speech (TTS) and voice conversion (VC), frequently become indistinguishable from genuine samples, posing challenges for individuals in discerning between real and synthetic content. This indistinguishability undermines trust in media, and the arbitrary cloning of personal voice signals presents significant challenges to privacy and security. In the field of deepfake audio detection, the majority of models achieving higher detection accuracy currently employ self-supervised pre-trained models. However, with the ongoing development of deepfake audio generation algorithms, maintaining high discrimination accuracy against new algorithms grows more challenging. To enhance the sensitivity of deepfake audio features, we propose a deepfake audio detection model that incorporates an SLS (Sensitive Layer Selection) module. Specifically, utilizing the pre-trained XLS-R enables our model to extract diverse audio features from its various layers, each providing distinct discriminative information. Utilizing the SLS classifier, our model captures sensitive contextual information across different layer levels of audio features, effectively employing this information for fake audio detection. Experimental results show that our method achieves state-of-the-art (SOTA) performance on both the ASVspoof 2021 DF and In-the-Wild datasets, with a specific Equal Error Rate (EER) of 1.92% on the ASVspoof 2021 DF dataset and 7.46% on the In-the-Wild dataset. Codes and data can be found at https://github.com/QiShanZhang/SLSforADD.
This paper studies the task of speech reconstruction from ultrasound tongue images and optical lip videos recorded in a silent speaking mode, where people only activate their intra-oral and extra-oral articulators without producing real speech. This task falls under the umbrella of articulatory-to-acoustic (A2A) conversion and may also be referred to as a silent speech interface. To overcome the domain discrepancy between silent and standard vocalized articulation, we introduce a novel pseudo target generation strategy. It integrates the text modality to align with articulatory movements, thereby guiding the generation of pseudo acoustic features for supervised training on speech reconstruction from silent articulation. Furthermore, we propose to employ a denoising diffusion probabilistic model as the fundamental architecture for the A2A conversion task and train the model using a combined training approach with the generated pseudo acoustic features. Experiments show that our proposed method significantly improves the intelligibility and naturalness of the reconstructed speech in the silent speaking mode compared to all baseline methods. Specifically, the word error rate of the reconstructed speech decreases by approximately 5% when measured using an automatic speech recognition engine for intelligibility assessment, and the subjective mean opinion score for naturalness improves by 0.14. Moreover, analytical experiments reveal that the proposed pseudo target generation strategy can generate pseudo acoustic features that synchronize better with articulatory movements than previous strategies. Samples are available at our project page.
Co-Speech gesture generation encounters challenges with imbalanced, long-tailed gesture distributions. While recent methods typically address this by employing Vector Quantized Variational Autoencoder (VQ-VAE), encode gestures into a codebook and classify codebook indices based on audio or text cues. However, due to the imbalanced, the codebook classification tends to bias towards majority gestures, neglecting semantically rich minority gestures. To address this, this paper proposes the Entropy-Guided Co-Speech Gesture Generation (EGGesture). EGGesture leverages an Entropy-Guided VQ-VAE to jointly optimizes the distribution of codebook indices and adjusts loss weights for codebook index classification, which consists of a) A differentiable approach for entropy computation using Gumbel-Softmax and cosine similarity, facilitating online codebook distribution optimization, and b) a strategy that utilizes computed codebook entropy to collaboratively guide the classification loss weighting. These designs enable the dynamic refinement of the codebook utilization, striking a balance between the quality of the learned gesture representation and the accuracy of the classification phase. Experiments on the Trinity and BEAT datasets demonstrate EGGesture's state-of-the-art performance both qualitatively and quantitatively.
Multimodal Emotion Recognition in Conversations aims to understand the human emotion of each utterance in a conversation from different types of data, such as speech and text. Previous works mainly focus on either complex unimodal feature extraction or sophisticated fusion techniques as general multimodal classification tasks do. However, they ignore the process of human perception, neglecting various levels of emotional features within each modality and disregarding the unique contributions of different modalities for emotion recognition. To address these issues, we propose a more cognitive-aligned multimodal fusion framework, namely DQ-Former. Specifically, DQ-Former utilizes a small set of learnable query tokens to collate and condense various granularities of emotion cues embedded at different layers of pre-trained unimodal models. Subsequently, it integrates these emotional features from different modalities with dynamic modality priorities at each intermediate fusion layer. This process enables explicit and effective fusion of different levels of information from diverse modalities. Extensive experiments on MELD and IEMOCAP datasets validate the effectiveness of DQ-Former. Our results show that the proposed method achieves a robust and interpretable multimodal representation for emotion recognition.
RAVSS: Robust Audio-Visual Speech Separation in Multi-Speaker Scenarios with Missing Visual Cues
PDF ↗While existing Audio-Visual Speech Separation (AVSS) methods primarily concentrate on the audio-visual fusion strategy for two-speaker separation, they demonstrate a severe performance drop in the multi-speaker separation scenarios. Typically, AVSS methods employ guiding videos to sequentially isolate individual speakers from the given audio mixture, resulting in notable missing and noisy parts across various segments of the separated speech. In this study, we propose a simultaneous multi-speaker separation framework that can facilitate the concurrent separation of multiple speakers within a singular process. We introduce speaker-wise interactions to establish distinctions and correlations among speakers. Experimental results on the VoxCeleb2 and LRS3 datasets demonstrate that our method achieves state-of-the-art performance in separating mixtures with 2, 3, 4, and 5 speakers, respectively. Additionally, our model can utilize speakers with complete audio-visual information to mitigate other visual-deficient speakers, thereby enhancing its resilience to missing visual cues. We also conduct experiments where visual information for specific speakers is entirely absent or visual frames are partially missing. The results demonstrate that our model consistently outperforms others, exhibiting the smallest performance drop across all settings involving 2, 3, 4, and 5 speakers.
Low resource of parallel data is the key challenge of accent conversion(AC) problem in which both the pronunciation units and prosody pattern need to be converted. We propose a two-stage generative framework "convert-and-speak" in which the conversion is only operated on the semantic token level and the speech is synthesized conditioned on the converted semantic token with a speech generative model in target accent domain. The decoupling design enables the "speaking" module to use massive amount of target accent speech and relieves the parallel data required for the "conversion" module. Conversion with the bridge of semantic token also relieves the requirement for the data with text transcriptions and unlocks the usage of language pre-training technology to further efficiently reduce the need of parallel accent speech data. To reduce the complexity and latency of "speaking", a single-stage AR generative model is designed to achieve good quality as well as lower computation cost. Experiments on Indian-English to general American-English conversion show that the proposed framework achieves state-of-the-art performance in accent similarity, speech quality, and speaker maintenance with only 15 minutes of weakly parallel data which is not constrained to the same speaker. Extensive experimentation with diverse accent types suggests that this framework possesses a high degree of adaptability, making it readily scalable to accommodate other accents with low-resource data. Audio samples are available at https://www.microsoft.com/en-us/research/project/convert-and-speak-zero-shot-accent-conversion-with-minimumsupervision/.
Conversational Speech Synthesis (CSS) aims to express a target utterance with the proper speaking style in a user-agent conversation setting. Existing CSS methods employ effective multi-modal context modeling techniques to achieve empathy understanding and expression. However, they often need to design complex network architectures and meticulously optimize the modules within them. In addition, due to the limitations of small-scale datasets containing scripted recording styles, they often fail to simulate real natural conversational styles. To address the above issues, we propose a novel generative expressive CSS system, termed GPT-Talker.We transform the multimodal information of the multi-turn dialogue history into discrete token sequences and seamlessly integrate them to form a comprehensive user-agent dialogue context. Leveraging the power of GPT, we predict the token sequence, that includes both semantic and style knowledge, of response for the agent. After that, the expressive conversational speech is synthesized by the conversation-enriched VITS to deliver feedback to the user.Furthermore, we propose a large-scale Natural CSS Dataset called NCSSD, that includes both naturally recorded conversational speech in improvised styles and dialogues extracted from TV shows. It encompasses both Chinese and English languages, with a total duration of 236 hours. We conducted comprehensive experiments on the reliability of the NCSSD and the effectiveness of our GPT-Talker. Both subjective and objective evaluations demonstrate that our model outperforms other state-of-the-art CSS systems significantly in terms of naturalness and expressiveness. The Code, Dataset, and Pre-trained Model are available at: https://github.com/AI-S2-Lab/GPT-Talker.
In the burgeoning field of Audio-Visual Speech Recognition (AVSR), extant research has predominantly concentrated on the training paradigms tailored for high-quality resources. However, owing to the challenges inherent in real-world data collection, audio-visual data are frequently affected by modality-distortion, which encompasses audio-visual asynchrony, video noise and audio noise. The recognition accuracy of existing AVSR method is significantly compromised when multiple modality-distortion coexist in low-resource data. In light of the above challenges, we propose PCD: cluster-Prompt with Contrastive Decomposition, a robust framework for modality-distortion speech recognition, specifically devised to transpose the pre-trained knowledge from high-resource domain to the targeted domain by leveraging contrast-augmented prompts. In contrast to previous studies, we take into consideration the possibility of various types of distortion in both the audio and visual modalities. Concretely, we design bespoke prompts to delineate each modality-distortion, guiding the model to achieve speech recognition applicable to various distortion scenarios with quite few learnable parameters. To materialize the prompt mechanism, we employ multiple cluster-based strategies that better suits the pre-trained audio-visual model. Additionally, we design a contrastive decomposition mechanism to restrict the explicit relationships among various modality conditions, given their shared task knowledge and disparate modality priors. Extensive results on LRS2 dataset demonstrate that PCD achieves state-of-the-art performance for audio-visual speech recognition under the constraints of distorted resources. Code is available at https://github.com/ballooncatt/PCD.