Search

Recently, the application of diffusion models has facilitated the significant development of speech and audio generation. Nevertheless, the quality of samples generated by diffusion models still needs improvement. And the effectiveness of the method is accompanied by the extensive number of sampling steps, leading to an extended synthesis time necessary for generating high-quality audio. Previous Text-to-Audio (TTA) methods mostly used diffusion models in the latent space for audio generation. In this paper, we explore the integration of the Flow Matching (FM) model into the audio latent space for audio generation. The FM is an alternative simulation-free method that trains continuous normalization flows (CNF) based on regressing vector fields. We demonstrate that our model significantly enhances the quality of generated audio samples, achieving better performance than prior models. Moreover, it reduces the number of inference steps to ten steps almost without sacrificing performance., Comment: Accepted at Interspeech2024

Recently, x-vector has been a successful and popular approach for speaker verification, which employs a time delay neural network (TDNN) and statistics pooling to extract speaker characterizing embedding from variable-length utterances. Improvement upon the x-vector has been an active research area, and enormous neural networks have been elaborately designed based on the x-vector, eg, extended TDNN (E-TDNN), factorized TDNN (F-TDNN), and densely connected TDNN (D-TDNN). In this work, we try to identify the optimal architectures from a TDNN based search space employing neural architecture search (NAS), named SpeechNAS. Leveraging the recent advances in the speaker recognition, such as high-order statistics pooling, multi-branch mechanism, D-TDNN and angular additive margin softmax (AAM) loss with a minimum hyper-spherical energy (MHE), SpeechNAS automatically discovers five network architectures, from SpeechNAS-1 to SpeechNAS-5, of various numbers of parameters and GFLOPs on the large-scale text-independent speaker recognition dataset VoxCeleb1. Our derived best neural network achieves an equal error rate (EER) of 1.02% on the standard test set of VoxCeleb1, which surpasses previous TDNN based state-of-the-art approaches by a large margin. Code and trained weights are in https://github.com/wentaozhu/speechnas.git, Comment: 8 pages, 3 figures, 3 tables. Accepted by ASRU2021

The audio source separation tasks, such as speech enhancement, speech separation, and music source separation, have achieved impressive performance in recent studies. The powerful modeling capabilities of deep neural networks give us hope for more challenging tasks. This paper launches a new multi-task audio source separation (MTASS) challenge to separate the speech, music, and noise signals from the monaural mixture. First, we introduce the details of this task and generate a dataset of mixtures containing speech, music, and background noises. Then, we propose an MTASS model in the complex domain to fully utilize the differences in spectral characteristics of the three audio signals. In detail, the proposed model follows a two-stage pipeline, which separates the three types of audio signals and then performs signal compensation separately. After comparing different training targets, the complex ratio mask is selected as a more suitable target for the MTASS. The experimental results also indicate that the residual signal compensation module helps to recover the signals further. The proposed model shows significant advantages in separation performance over several well-known separation models.

We study entanglement witness and present a construction of entanglement witnesses in terms of the symmetric informationally complete measurements (SIC-POVM). The capability of our witness is shown by some examples and it can be found this witness detects more entanglement than previous separability method given also by SIC-POVM. What's more, comparing with the dimension dependence of SIC-POVM, we also give the entanglement witnesses can be always constructed from general symmetric informationally complete measurements (GSIC-POVM).

Optical detection of spin state has been widely used for the solid state spin qubit in the application of quantum information processing. The signal contrast determines the accuracy of quantum state manipulation, sensitivity of quantum sensing and resolution of quantum imaging. Here, we demonstrated a time-gated fluorescence detection method for enhancing the spin state signal contrast of nitrogen vacancy (NV) center in diamond. By adjusting the delay between time gate and the excitation laser pulse, we improved both the signal contrast and signal-to-noise ratio for NV spin detection. An enhancement ratio of 1.86 times was reached for the signal contrast. Utilizing the time-gated fluorescence detection, we further demonstrated a high contrast quantum imaging of nanoparticle's stray magnetic field. Without any additional manipulation of the quantum state, we expect that this method can be used to improve the performance of various applications with NV center.

The dispersion and coordination of aluminium species on the surface of silica-alumina based materials are essential for controlling their catalytic activity and selectivity. Al(IV) and Al(VI) are two common coordinations of Al species in the silica network and alumina phase, respectively. Al(V) is rare in nature and was found hitherto only in the alumina phase or interfaces containing alumina, a behavior which negatively affects the dispersion, population, and accessibility of Al(V) species on the silica-alumina surface. This constraint has limited the development of silica-alumina based catalysts, particularly because Al(V) had been confirmed to act as a highly active center for acid reactions and single-atom catalysts. Here, we report the direct observation of high population of homogenously dispersed Al(V) species in amorphous silica-alumina in the absence of any bulk alumina phase, by high resolution TEM/EDX and high magnetic-field MAS NMR. Solid-state 27Al multi-quantum MAS NMR experiments prove unambiguously that most of the Al(V) species formed independently from the alumina phase and are accessible on the surface for guest molecules. These species are mainly transferred to Al(VI) species with partial formation of Al(IV) species after adsorption of water. The NMR chemical shifts and their coordination transformation with and without water adsorption are matching that obtained in DFT calculations of the predicted clusters. The discovery presented in this study not only provides fundamental knowledge of the nature of aluminum coordination, but also paves the way for developing highly efficient catalysts., Comment: 13 pages, 4 figures, 3 tables

We have studied the exotic superfluid phases of degenerated Fermi gases with spin-orbit coupling in a mixed dimensional system, where the motion of atoms are free in the x-direction and the tunneling between nearest tubes in the y-direction is permitted. Using the mean-field method, we obtain the phase diagrams of the system during the dimensional crossover between quasi one dimension to quasi two dimension. We find the existence of the topological state and Majorana edge mode in the weak tunneling case, and a rich phase diagram including two kinds of nodal superfluid phase and gapped superfluid phase in the opposite case. Our results show that topological pairing is favoured in quasi one dimension while nodal pairing state is favoured in quasi two dimension.

This review examines 105 empirical studies that investigate students' views of the nature of science (VNOS), effects of curricular interventions on changing students' VNOS, and relations between VNOS and demographics, majors, and learning of science. The reviewed studies can be categorized into three theoretical frameworks: the unidimension, the multidimension, and the argumentative resource frameworks. Each framework is reviewed first with regard to its theoretical foundation, methods of data collection and analysis, and the respective findings. This is followed by a critical discussion on the methodological issues and the strengths and limitations of the framework. Potential directions regarding theoretical framework, methodology, and pedagogy are suggested in the future directions section. For example, this review proposes the shift to argumentative resource framework for VNOS research and a refinement for judging the "sophistication" of student VNOS. The role of inquiry in student VNOS is also discussed. (Contains 1 table.)