Abstract:Zero-shot voice conversion (VC) aims to transform the source speaker timbre into an arbitrary unseen one without altering the original speech content.While recent advancements in zero-shot VC methods have shown remarkable progress, there still remains considerable potential for improvement in terms of improving speaker similarity and speech naturalness.In this paper, we propose Takin-VC, a novel zero-shot VC framework based on jointly hybrid content and memory-augmented context-aware timbre modeling to tackle this challenge. Specifically, an effective hybrid content encoder, guided by neural codec training, that leverages quantized features from pre-trained WavLM and HybridFormer is first presented to extract the linguistic content of the source speech. Subsequently, we introduce an advanced cross-attention-based context-aware timbre modeling approach that learns the fine-grained, semantically associated target timbre features. To further enhance both speaker similarity and real-time performance, we utilize a conditional flow matching model to reconstruct the Mel-spectrogram of the source speech. Additionally, we advocate an efficient memory-augmented module designed to generate high-quality conditional target inputs for the flow matching process, thereby improving the overall performance of the proposed system. Experimental results demonstrate that the proposed Takin-VC method surpasses state-of-the-art zero-shot VC systems, delivering superior performance in terms of both speech naturalness and speaker similarity.
Abstract:With the advent of the big data and large language model era, zero-shot personalized rapid customization has emerged as a significant trend. In this report, we introduce Takin AudioLLM, a series of techniques and models, mainly including Takin TTS, Takin VC, and Takin Morphing, specifically designed for audiobook production. These models are capable of zero-shot speech production, generating high-quality speech that is nearly indistinguishable from real human speech and facilitating individuals to customize the speech content according to their own needs. Specifically, we first introduce Takin TTS, a neural codec language model that builds upon an enhanced neural speech codec and a multi-task training framework, capable of generating high-fidelity natural speech in a zero-shot way. For Takin VC, we advocate an effective content and timbre joint modeling approach to improve the speaker similarity, while advocating for a conditional flow matching based decoder to further enhance its naturalness and expressiveness. Last, we propose the Takin Morphing system with highly decoupled and advanced timbre and prosody modeling approaches, which enables individuals to customize speech production with their preferred timbre and prosody in a precise and controllable manner. Extensive experiments validate the effectiveness and robustness of our Takin AudioLLM series models. For detailed demos, please refer to https://takinaudiollm.github.io.
Abstract:Speaker anonymization is an effective privacy protection solution designed to conceal the speaker's identity while preserving the linguistic content and para-linguistic information of the original speech. While most prior studies focus solely on a single language, an ideal speaker anonymization system should be capable of handling multiple languages. This paper proposes MUSA, a Multi-lingual Speaker Anonymization approach that employs a serial disentanglement strategy to perform a step-by-step disentanglement from a global time-invariant representation to a temporal time-variant representation. By utilizing semantic distillation and self-supervised speaker distillation, the serial disentanglement strategy can avoid strong inductive biases and exhibit superior generalization performance across different languages. Meanwhile, we propose a straightforward anonymization strategy that employs empty embedding with zero values to simulate the speaker identity concealment process, eliminating the need for conversion to a pseudo-speaker identity and thereby reducing the complexity of speaker anonymization process. Experimental results on VoicePrivacy official datasets and multi-lingual datasets demonstrate that MUSA can effectively protect speaker privacy while preserving linguistic content and para-linguistic information.
Abstract:Interpreting complex deep networks, notably pre-trained vision-language models (VLMs), is a formidable challenge. Current Class Activation Map (CAM) methods highlight regions revealing the model's decision-making basis but lack clear saliency maps and detailed interpretability. To bridge this gap, we propose DecomCAM, a novel decomposition-and-integration method that distills shared patterns from channel activation maps. Utilizing singular value decomposition, DecomCAM decomposes class-discriminative activation maps into orthogonal sub-saliency maps (OSSMs), which are then integrated together based on their contribution to the target concept. Extensive experiments on six benchmarks reveal that DecomCAM not only excels in locating accuracy but also achieves an optimizing balance between interpretability and computational efficiency. Further analysis unveils that OSSMs correlate with discernible object components, facilitating a granular understanding of the model's reasoning. This positions DecomCAM as a potential tool for fine-grained interpretation of advanced deep learning models. The code is avaible at https://github.com/CapricornGuang/DecomCAM.
Abstract:The continuous evolution of pre-trained speech models has greatly advanced Speech Emotion Recognition (SER). However, there is still potential for enhancement in the performance of these methods. In this paper, we present GMP-ATL (Gender-augmented Multi-scale Pseudo-label Adaptive Transfer Learning), a novel HuBERT-based adaptive transfer learning framework for SER. Specifically, GMP-ATL initially employs the pre-trained HuBERT, implementing multi-task learning and multi-scale k-means clustering to acquire frame-level gender-augmented multi-scale pseudo-labels. Then, to fully leverage both obtained frame-level and utterance-level emotion labels, we incorporate model retraining and fine-tuning methods to further optimize GMP-ATL. Experiments on IEMOCAP show that our GMP-ATL achieves superior recognition performance, with a WAR of 80.0\% and a UAR of 82.0\%, surpassing state-of-the-art unimodal SER methods, while also yielding comparable results with multimodal SER approaches.
Abstract:Speaker-attributed automatic speech recognition (SA-ASR) improves the accuracy and applicability of multi-speaker ASR systems in real-world scenarios by assigning speaker labels to transcribed texts. However, SA-ASR poses unique challenges due to factors such as speaker overlap, speaker variability, background noise, and reverberation. In this study, we propose PP-MeT system, a real-world personalized prompt based meeting transcription system, which consists of a clustering system, target-speaker voice activity detection (TS-VAD), and TS-ASR. Specifically, we utilize target-speaker embedding as a prompt in TS-VAD and TS-ASR modules in our proposed system. In constrast with previous system, we fully leverage pre-trained models for system initialization, thereby bestowing our approach with heightened generalizability and precision. Experiments on M2MeT2.0 Challenge dataset show that our system achieves a cp-CER of 11.27% on the test set, ranking first in both fixed and open training conditions.
Abstract:Style voice conversion aims to transform the style of source speech to a desired style according to real-world application demands. However, the current style voice conversion approach relies on pre-defined labels or reference speech to control the conversion process, which leads to limitations in style diversity or falls short in terms of the intuitive and interpretability of style representation. In this study, we propose PromptVC, a novel style voice conversion approach that employs a latent diffusion model to generate a style vector driven by natural language prompts. Specifically, the style vector is extracted by a style encoder during training, and then the latent diffusion model is trained independently to sample the style vector from noise, with this process being conditioned on natural language prompts. To improve style expressiveness, we leverage HuBERT to extract discrete tokens and replace them with the K-Means center embedding to serve as the linguistic content, which minimizes residual style information. Additionally, we deduplicate the same discrete token and employ a differentiable duration predictor to re-predict the duration of each token, which can adapt the duration of the same linguistic content to different styles. The subjective and objective evaluation results demonstrate the effectiveness of our proposed system.
Abstract:Contrastive learning based pretraining methods have recently exhibited impressive success in diverse fields. In this paper, we propose GEmo-CLAP, a kind of efficient gender-attribute-enhanced contrastive language-audio pretraining (CLAP) model for speech emotion recognition. To be specific, we first build an effective emotion CLAP model Emo-CLAP for emotion recognition, utilizing various self-supervised learning based pre-trained models. Then, considering the importance of the gender attribute in speech emotion modeling, two GEmo-CLAP approaches are further proposed to integrate the emotion and gender information of speech signals, forming more reasonable objectives. Extensive experiments on the IEMOCAP corpus demonstrate that our proposed two GEmo-CLAP approaches consistently outperform the baseline Emo-CLAP with different pre-trained models, while also achieving superior recognition performance compared with other state-of-the-art methods.
Abstract:The emergence of cross-modal foundation models has introduced numerous approaches grounded in text-image retrieval. However, on some domain-specific retrieval tasks, these models fail to focus on the key attributes required. To address this issue, we propose a self-enhancement framework, A^{3}R, based on the CLIP-ViT/G-14, one of the largest cross-modal models. First, we perform an Attribute Augmentation strategy to enrich the textual description for fine-grained representation before model learning. Then, we propose an Adaption Re-ranking method to unify the representation space of textual query and candidate images and re-rank candidate images relying on the adapted query after model learning. The proposed framework is validated to achieve a salient improvement over the baseline and other teams' solutions in the cross-modal image retrieval track of the 1st foundation model challenge without introducing any additional samples. The code is available at \url{https://github.com/CapricornGuang/A3R}.
Abstract:Interpretation of deep learning remains a very challenging problem. Although the Class Activation Map (CAM) is widely used to interpret deep model predictions by highlighting object location, it fails to provide insight into the salient features used by the model to make decisions. Furthermore, existing evaluation protocols often overlook the correlation between interpretability performance and the model's decision quality, which presents a more fundamental issue. This paper proposes a new two-stage interpretability method called the Decomposition Class Activation Map (Decom-CAM), which offers a feature-level interpretation of the model's prediction. Decom-CAM decomposes intermediate activation maps into orthogonal features using singular value decomposition and generates saliency maps by integrating them. The orthogonality of features enables CAM to capture local features and can be used to pinpoint semantic components such as eyes, noses, and faces in the input image, making it more beneficial for deep model interpretation. To ensure a comprehensive comparison, we introduce a new evaluation protocol by dividing the dataset into subsets based on classification accuracy results and evaluating the interpretability performance on each subset separately. Our experiments demonstrate that the proposed Decom-CAM outperforms current state-of-the-art methods significantly by generating more precise saliency maps across all levels of classification accuracy. Combined with our feature-level interpretability approach, this paper could pave the way for a new direction for understanding the decision-making process of deep neural networks.