Transformer has achieved competitive performance against state-of-the-art end-to-end models in automatic speech recognition (ASR), and requires significantly less training time than RNN-based models.The original Transformer, with encoderdecoder architecture, is only suitable for offline ASR.It relies on an attention mechanism to learn alignments, and encodes input audio bidirectionally.The high computation cost of Transformer decoding also limits its use in production streaming systems.To make Transformer suitable for streaming ASR, we explore Transducer framework as a streamable way to learn alignments.For audio encoding, we apply unidirectional Transformer with interleaved convolution layers.The interleaved convolution layers are used for modeling future context which is important to performance.To reduce computation cost, we gradually downsample acoustic input, also with the interleaved convolution layers.Moreover, we limit the length of history context in self-attention to maintain constant computation cost for each decoding step.We show that this architecture, named Conv-Transformer Transducer, achieves competitive performance on LibriSpeech dataset (3.6% WER on test-clean) without external language models.The performance is comparable to previously published streamable Transformer Transducer and strong hybrid streaming ASR systems, and is achieved with smaller look-ahead window (140 ms), fewer parameters and lower frame rate.

This research is aimed at the optimization of a two-dimensional (2D) empirical graph under a certain height and dark conditions for a UAV, using the bionic sonar system to replace the visual sensor’s BatSLAM mode and audio perceptual hash closed-loop detection. The BatSLAM model uses Sum of Absolute Difference (SAD) image processing methods to update the bionic sonar template. This method only judges whether the appearance of the two cochlear images is consistent and does not have geometric processing and feature extraction. Because the cochlear images produce various noises during the acquisition and transmission, there are some differences in cochlear maps obtained at the same position, which can lead to the distortion of the constructed empirical map. In this research, an audio perceptual hash closed-loop detection algorithm is developed to extract features of cochlea. It considers both the appearance and the energy difference between adjacent bands to improve the accuracy of closed-loop detection, thus solving the distortion problem and improving the experience map. The simulation experiment shows that the improved BatSLAM model based on the audio perceptual hash closed-loop detection can improve the 2D experience map for UAV under certain height and dark conditions, through improving the accuracy of the closed-loop detection to solve the distortion problem and thus implementing the optimization of the experience graph.

Mosaics are widely used for surface decoration to produce appealing visual effects. We present a method for synthesizing digital surface mosaics with irregularly shaped tiles, which are a type of tiles often used for mosaics design. Our method employs both continuous optimization and combinatorial optimization to improve tile arrangement. In the continuous optimization step, we iteratively partition the base surface into approximate Voronoi regions of the tiles and optimize the positions and orientations of the tiles to achieve a tight fit. Combination optimization performs tile permutation and replacement to further increase surface coverage and diversify tile selection. The alternative applications of these two optimization steps lead to rich combination of tiles and high surface coverage. We demonstrate the effectiveness of our solution with extensive experiments and comparisons.

We present a new method for modeling discrete constant mean curvature (CMC) surfaces, which arise frequently in nature and are highly demanded in architecture and other engineering applications. Our method is based on a novel use of the CVT ( centroidal Voronoi tessellation ) optimization framework. We devise a CVT-CMC energy function defined as a combination of an extended CVT energy and a volume functional. We show that minimizing the CVT-CMC energy is asymptotically equivalent to minimizing mesh surface area with a fixed volume, thus defining a discrete CMC surface. The CVT term in the energy function ensures high mesh quality throughout the evolution of a CMC surface in an interactive design process for form finding. Our method is capable of modeling CMC surfaces with fixed or free boundaries and is robust with respect to input mesh quality and topology changes. Experiments show that the new method generates discrete CMC surfaces of improved mesh quality over existing methods.