Related papers: Optimizing Audio Compression Through Entropy-Contr…

In this paper, we perform an in-depth study of how data augmentation techniques improve synthetic or spoofed audio detection. Specifically, we propose methods to deal with channel variability, different audio compressions, different…

We study the maximum achievable differential entropy at the output of a system assigning to each input X the sum X+N, with N a given noise with probability law absolutely continuous with respect to the Lebesgue measure and where the input…

Entropy serves as a critical metric for measuring the diversity of outputs generated by large language models (LLMs), providing valuable insights into their exploration capabilities. While recent studies increasingly focus on monitoring and…

Diffusion models produce high-fidelity speech but are inefficient for real-time use due to long denoising steps and challenges in modeling intonation and rhythm. To improve this, we propose Diffusion Loss-Guided Policy Optimization (DLPO),…

Especially in lossless image coding the obtainable compression ratio strongly depends on the amount of noise included in the data as all noise has to be coded, too. Different approaches exist for lossless image coding. We analyze the…

Data augmentation is a key tool for improving the performance of deep networks, particularly when there is limited labeled data. In some fields, such as computer vision, augmentation methods have been extensively studied; however, for…

Discrete speech representation learning has recently attracted increasing interest in both acoustic and semantic modeling. Existing approaches typically encode 16 kHz waveforms into discrete tokens at a rate of 25 or 50 tokens per second.…

Diffusion models have achieved state-of-the-art synthesis quality on both visual and audio tasks, and recent works further adapt them to textual data by diffusing on the embedding space. In this paper, we conduct systematic studies of the…

The goal of this paper is to adapt speaker embeddings for solving the problem of speaker diarisation. The quality of speaker embeddings is paramount to the performance of speaker diarisation systems. Despite this, prior works in the field…

Data-enabled predictive control (DeePC) leverages system measurements in characterizing system dynamics for optimal control. The performance of DeePC relies on optimizing its hyperparameters, especially in noisy systems where the optimal…

Signal to noise ratio is key to any measurement. Recent progress in semi/super-conductor technology have pushed the signal detection sensitivity to the ultimate quantum level, but the noise issue remains largely untouched and, in many…

Compressing word embeddings is important for deploying NLP models in memory-constrained settings. However, understanding what makes compressed embeddings perform well on downstream tasks is challenging---existing measures of compression…

Recent end-to-end deep neural networks for disparity regression have achieved the state-of-the-art performance. However, many well-acknowledged specific properties of disparity estimation are omitted in these deep learning algorithms.…

Inspired by recent developments in neural speech coding and diffusion-based language modeling, we tackle speech enhancement by modeling the conditional distribution of clean speech codes given noisy speech codes using absorbing discrete…

Deep learning models have been widely used in commercial acoustic systems in recent years. However, adversarial audio examples can cause abnormal behaviors for those acoustic systems, while being hard for humans to perceive. Various…

Symmetry-adapted distortion modes provide a natural way to describe distorted structures derived from higher-symmetry parent phases. Structural refinements using symmetry-mode amplitudes as fit variables have been used for at least 10 years…

Differential privacy provides the first theoretical foundation with provable privacy guarantee against adversaries with arbitrary prior knowledge. The main idea to achieve differential privacy is to inject random noise into statistical…

Automated audio captioning aims to use natural language to describe the content of audio data. This paper presents an audio captioning system with an encoder-decoder architecture, where the decoder predicts words based on audio features…

The recently developed pitch-controllable text-to-speech (TTS) model, i.e. FastPitch, was conditioned for the pitch contours. However, the quality of the synthesized speech degraded considerably for pitch values that deviated significantly…

Randomized (dithered) quantization is a method capable of achieving white reconstruction error independent of the source. Dithered quantizers have traditionally been considered within their natural setting of uniform quantization. In this…