Yong-Ju Lee

This paper reports labeler agreement in the transcription of Korean prosody using Korean ToBI (K-ToBI) [9]. Twenty utterances representing five different types of speech were produced by 18 speakers and transcribed by 21 labelers differing in their levels of experience with K-ToBI. Following the stringent metric used for English ToBI evaluation [14,12], consistency was measured in terms of the number of transcriber pairs agreeing on the labeling of each particular word. The results show that for tonal transcriptions of the 32,130 transcriber-pair-words, agreement was 77 % for the type of boundaries at the end of each word (i.e., word, AP, or IP), 78% for AP boundaries, and 91 % for IP boundaries. For break indices, the agreement score for exact matching in the labeling was 59%, 69 % when relaxing the presence/absence of diacritics, and 99 % when relaxing within +/-1 level. In sum, the data confirm that the conventions of K-ToBI are adequate, easy to learn, and can be reliably used for research in Korean prosody and for large-scale prosodic annotation in speech databases.

Abstract Advances in AI are transforming scientific discovery, yet spatial biology, a field that deciphers the molecular organization within tissues, remains constrained by labor-intensive workflows. Here, we present SpatialAgent, a fully autonomous AI agent dedicated for spatial-biology research. SpatialAgent integrates large language models with dynamic tool execution and adaptive reasoning. SpatialAgent spans the entire research pipeline, from experimental design to multimodal data analysis and hypothesis generation. Tested on multiple datasets comprising two million cells from human brain, heart, and a mouse colon colitis model, SpatialAgent’s performance surpassed the best computational methods, matched or outperformed human scientists across key tasks, and scaled across tissues and species. By combining autonomy with human collaboration, SpatialAgent establishes a new paradigm for AI-driven discovery in spatial biology.

In recent days, unmanned aerial vehicles (UAVs) becomes more familiar because of its versatility, automation abilities, and low cost. Dynamic scene classification gained significant interest among the UAV-based surveillance systems, e.g., high-voltage power line and forest fire monitoring, which facilitate the object detection, tracking process and drastically enhances the outcome of visual surveillance. This paper proposes a new optimal deep learning-based scene classification model captured by UAVs. The proposed model involves a residual network-based features extraction (RNBFE) which extracts features from the diverse convolution layers of a deep residual network. In addition, the several parameters in RNBFE lead to many configuration errors due to manual parameter tuning. So, self-adaptive global best harmony search (SGHS) algorithm is employed for tuning the parameters of the RNBFE. The resultant feature vectors undergo classification by the use of latent variable support vector machine (LVSVM) model. The presented optimal RNBFE (ORNBFE) model has been tested using two open access datasets namely UC Merced (UCM) Land Use Dataset and WHU-RS Dataset. The presented technique attains maximum scene classification accuracy over the other recently proposed methods.