Learning Semantic Information from Raw Audio Signal Using Both Contextual and Phonetic Representations

We propose a framework to learn semantics from raw audio signals using two types of representations, encoding contextual and phonetic information respectively. Specifically, we introduce a speech-to-unit processing pipeline that captures two types of representations with different time resolutions. For the language model, we adopt a dual-channel architecture to incorporate both types of representation. We also present new training objectives, masked context reconstruction and masked context prediction, that push models to learn semantics effectively. Experiments on the sSIMI metric of Zero Resource Speech Benchmark 2021 and Fluent Speech Command dataset show our framework learns semantics better than models trained with only one type of representation.

EnCLAP: Combining Neural Audio Codec and Audio-Text Joint Embedding for Automated Audio Captioning

We propose EnCLAP, a novel framework for automated audio captioning. EnCLAP employs two acoustic representation models, EnCodec and CLAP, along with a pretrained language model, BART. We also introduce a new training objective called masked codec modeling that improves acoustic awareness of the pretrained language model. Experimental results on AudioCaps and Clotho demonstrate that our model surpasses the performance of baseline models. Source code will be available at https://github.com/jaeyeonkim99/EnCLAP . An online demo is available at https://huggingface.co/spaces/enclap-team/enclap .

Language-driven Object Fusion into Neural Radiance Fields with Pose-Conditioned Dataset Updates

Neural radiance field is an emerging rendering method that generates high-quality multi-view consistent images from a neural scene representation and volume rendering. Although neural radiance field-based techniques are robust for scene reconstruction, their ability to add or remove objects remains limited. This paper proposes a new language-driven approach for object manipulation with neural radiance fields through dataset updates. Specifically, to insert a new foreground object represented by a set of multi-view images into a background radiance field, we use a text-to-image diffusion model to learn and generate combined images that fuse the object of interest into the given background across views. These combined images are then used for refining the background radiance field so that we can render view-consistent images containing both the object and the background. To ensure view consistency, we propose a dataset updates strategy that prioritizes radiance field training with camera views close to the already-trained views prior to propagating the training to remaining views. We show that under the same dataset updates strategy, we can easily adapt our method for object insertion using data from text-to-3D models as well as object removal. Experimental results show that our method generates photorealistic images of the edited scenes, and outperforms state-of-the-art methods in 3D reconstruction and neural radiance field blending.

PITS: Variational Pitch Inference without Fundamental Frequency for End-to-End Pitch-controllable TTS

Previous pitch-controllable text-to-speech (TTS) models rely on directly modeling fundamental frequency, leading to low variance in synthesized speech. To address this issue, we propose PITS, an end-to-end pitch-controllable TTS model that utilizes variational inference to model pitch. Based on VITS, PITS incorporates the Yingram encoder, the Yingram decoder, and adversarial training of pitch-shifted synthesis to achieve pitch-controllability. Experiments demonstrate that PITS generates high-quality speech that is indistinguishable from ground truth speech and has high pitch-controllability without quality degradation. Code and audio samples will be available at https://github.com/anonymous-pits/pits.

PointInverter: Point Cloud Reconstruction and Editing via a Generative Model with Shape Priors

In this paper, we propose a new method for mapping a 3D point cloud to the latent space of a 3D generative adversarial network. Our generative model for 3D point clouds is based on SP-GAN, a state-of-the-art sphere-guided 3D point cloud generator. We derive an efficient way to encode an input 3D point cloud to the latent space of the SP-GAN. Our point cloud encoder can resolve the point ordering issue during inversion, and thus can determine the correspondences between points in the generated 3D point cloud and those in the canonical sphere used by the generator. We show that our method outperforms previous GAN inversion methods for 3D point clouds, achieving state-of-the-art results both quantitatively and qualitatively. Our code is available at https://github.com/hkust-vgd/point_inverter.

Minimal Adversarial Examples for Deep Learning on 3D Point Clouds

With the recent developments of convolutional neural networks, deep learning for 3D point clouds has shown significant progress in various 3D scene understanding tasks including 3D object recognition. In a safety-critical environment, it is however not well understood how such neural networks are vulnerable to adversarial examples. In this work, we explore adversarial attacks for point cloud-based neural networks with a focus on real-world data. We propose a general formulation for adversarial point cloud generation via $\ell_0$-norm optimisation. Our method generates adversarial examples by attacking the classification ability of the point cloud-based networks while considering the perceptibility of the examples and ensuring the minimum level of point manipulations. The proposed method is general and can be realised in different attack strategies. Experimental results show that our method achieves the state-of-the-art performance with 80\% of attack success rate while manipulating only about 4\% of the points. We also found that compared with synthetic data, real-world point cloud classification is more vulnerable to attacks.