Significant advances have been made in emotional speech synthesis technology; however, existing models still face challenges in achieving fine-grained emotion style control and simple yet precise emotion intensity regulation. To address these issues, we propose Easy-Control Emotion Text-to-Speech (ECE-TTS), a zero-shot TTS model built upon the F5-TTS architecture, simplifying emotion modeling while maintaining accurate control. ECE-TTS leverages pretrained emotion recognizers to extract Valence, Arousal, and Dominance (VAD) values, transforming them into Emotion-Adaptive Spherical Vectors (EASV) for precise emotion style representation. Emotion intensity modulation is efficiently realized via simple arithmetic operations on emotion vectors without introducing additional complex modules or training extra regression networks. Emotion style control experiments demonstrate that ECE-TTS achieves a Word Error Rate (WER) of 13.91%, an Aro-Val-Domin SIM of 0.679, and an Emo SIM of 0.594, surpassing GenerSpeech (WER = 16.34%, Aro-Val-Domin SIM = 0.627, Emo SIM = 0.563) and EmoSphere++ (WER = 15.08%, Aro-Val-Domin SIM = 0.656, Emo SIM = 0.578). Subjective Mean Opinion Score (MOS) evaluations (1–5 scale) further confirm improvements in speaker similarity (3.93), naturalness (3.98), and emotional expressiveness (3.94). Additionally, emotion intensity control experiments demonstrate smooth and precise modulation across varying emotional strengths. These results validate ECE-TTS as a highly effective and practical solution for high-quality, emotion-controllable speech synthesis.

Radio Frequency Fingerprint Identification (RFFI) is a promising device authentication technique that utilizes inherent hardware flaws in transmitters to achieve device identification, thus effectively maintaining the security of the Internet of Things (IoT). However, time-varying channels degrade accuracy due to factors like device aging and environmental changes. To address this, we propose an RFFI method integrating Zero-Forcing (ZF) equalization and KAN-ResNet. Firstly, the Wi-Fi preamble signals under the IEEE 802.11 standard are Zero-Forcing equalized, so as to effectively reduce the interference of time-varying channels on RFFI. We then design a novel residual network, KAN-ResNet, which adds a KAN module on top of the traditional fully connected layer. The module combines the B-spline basis function and the traditional activation function Sigmoid Linear Unit (SiLU) to realize the nonlinear mapping of the complex function, which enhance the classification ability of the network for RFF features. In addition, to improve the generalization of the model, the grid of B-splines is dynamically updated and L1 regularization is introduced. Experiments show that on datasets collected 20 days apart, our method achieves 99.4% accuracy, reducing the error rate from 6.3% to 0.6%, outperforming existing models.

Speech anti-spoofing is a crucial aspect of speaker recognition systems and has received a great deal of attention in recent years. Deep neural networks have achieved satisfactory results in datasets with similar training and testing data distributions, but their generalization ability is limited in datasets with different distributions. In this paper, we proposed a novel dual-input hierarchical fusion network (HFN) to improve the generalization ability of our model. The network had two inputs (the original speech signal and the time-reversed signal), which increased the volume and diversity of the training data. The hierarchical fusion model (HFM) enabled more thorough fusion of information from different input levels and improved model performance by fusing the two inputs after speech feature extraction. We finally evaluated the results using the ASVspoof 2021 PA (Physical Access) dataset, and the proposed system achieved an Equal Error Rate (EER) of 24.46% and a minimum tandem Detection Cost Function (min t-DCF) of 0.6708 in the test set. Compared with the four baseline systems in the ASVspoof 2021 competition, the proposed system min t-DCF values were decreased by 28.9%, 31.0%, 32.6%, and 32.9%, and the EERs were decreased by 35.7%, 38.1%, 45.4%, and 49.7%, respectively.

As one of the most commonly used features, Mel-frequency cepstral coefficients (MFCCs) are less discriminative at high frequency. A novel technique, known as Deep scattering spectrum (DSS), addresses this issue and looks to preserve greater details. DSS feature has shown promise both on classification and recognition tasks. In this paper, we extend the use of DSS feature for acoustic scene classification task. Results on Detection and classification of acoustic scenes and events (DCASE) 2016 and 2017 show that DSS provided 4:8% and 17:4% relative improvements in accuracy over MFCC features, within a state-of-the-art time delay neural network framework.

The small receptive field and capacity of minimal neural networks limit their performance when using them to be the backbone of detectors. In this work, we find that the appearance feature of a generic face is discriminative enough for a tiny and shallow neural network to verify from the background. And the essential barriers behind us are 1) the vague definition of the face bounding box and 2) tricky design of anchor-boxes or receptive field. Unlike most top-down methods for joint face detection and alignment, the proposed KPNet detects small facial keypoints instead of the whole face by in a bottom-up manner. It first predicts the facial landmarks from a low-resolution image via the well-designed fine-grained scale approximation and scale adaptive soft-argmax operator. Finally, the precise face bounding boxes, no matter how we define it, can be inferred from the keypoints. Without any complex head architecture or meticulous network designing, the KPNet achieves state-of-the-art accuracy on generic face detection and alignment benchmarks with only $\sim1M$ parameters, which runs at 1000fps on GPU and is easy to perform real-time on most modern front-end chips.