In the era of human-robot symbiosis, endowing robots with emotional intelligence is essential for creating harmonious human-robot interactions and enabling them to fulfill social functions effectively. Although significant progress has been made in robotic emotion recognition, emotion generation remains underdeveloped, struggling to meet user-centered interaction demands in multi-turn, complex-task, and diverse-scenario settings. This often results in robotic behaviors that appear rigid and lack empathy. To address these challenges, this study proposes an Optimization Model for Robotic Artificial Emotion Generation (OMRAEG). Structured around a “theory–mechanism–method” framework, the model establishes a closed-loop optimization system encompassing method application, effectiveness evaluation, influencing factors, and feedback iteration. Specifically, the research integrates mainstream approaches and key technologies across four dimensions: facial expression synthesis, emotional dialogue generation, emotional speech synthesis, and emotional motion synthesis. Furthermore, it constructs a systematic evaluation indicator system and clarifies the fine-tuning role of application scenarios and development trends guiding technological evolution.

Prosody plays a fundamental role in human speech and communication, facilitating intelligibility and conveying emotional and cognitive states. Extracting accurate prosodic information from speech is vital for building assistive technology, such as controllable speech synthesis, speaking style transfer, and speech emotion recognition (SER). However, it is challenging to disentangle speaker-independent prosody representations since prosodic attributes, such as intonation, excessively entangle with speaker-specific attributes, e.g., pitch. In this article, we propose a novel model, called Diffsody, to disentangle and refine prosody representations: 1) to disentangle prosody representations, we leverage the expressive generative ability of a diffusion model by conditioning it on quantified semantic information and pretrained speaker embeddings. Additionally, a prosody encoder automatically learns prosody representations used for spectrogram reconstruction in an unsupervised fashion; and 2) to refine and learn speaker-invariant prosody representations, a scheduled gradient reversal layer (sGRL) is proposed and integrated into the prosody encoder of Diffsody. We extensively evaluate Diffsody through qualitative and quantitative means. t-SNE visualization and speaker verification experiments demonstrate the efficacy of the sGRL method in preventing speaker-specific information leakage. Experimental results on speaker-independent SER and automatic depression detection (ADD) tasks demonstrate that Diffsody can efficiently factorize speaker-independent prosody representations, resulting in a significant boost in SER and ADD. In addition, Diffsody synergistically integrates with the semantic representation model WavLM, which leads to a discernibly elevated performance, outperforming contemporary methods in both SER and ADD tasks. Furthermore, the Diffsody model exhibits promising potential for various practical applications, such as voice or style conversion. Some audio samples can be found on our https://leyuanqu.github.io/Diffsody/demo website.

Hyperspectral spectroscopy based on partial least squares regression (PLSR) is an effective tool for monitoring plant photosynthesis. Despite their wide applications, the robustness of PLSR models on tracing photosynthetic capacity, which varies considerably among different species and at different times, have been far less explored, leading to doubt about whether hyperspectral information can accurately predict the capacity across different species and temporal changes. Ordinary applications of PLSR generally make use of original or integer-order derivative transformed reflected spectra, but recent advances in spectral analysis have revealed that fractional-order derivative transformed spectra could provide more details of spectral signals. In this study, PLSR models based on fractional-order derivatives coupled with different wavelength selection methods were developed to evaluate whether photosynthetic parameters (Vcmax and Jmax) could be correctly predicted from reflectance spectra. The result indicated that the best PLSR models for the Vcmax and Jmax were obtained based on the sensitive wavelengths selected by stepwise regression using the fractional orders of 1.25 and 1.60, respectively. The optimal PLSR models were able to capture the temporal variabilities of Vcmax and Jmax with the R2of 0.62-0.94 and 0.65-0.85, for which the 1605-1845 nm region was consistently used. Meanwhile, these PLSR models have the ability to capture the variations in different species, plant functional types, and biomes. The findings of this study demonstrate that leaf spectra can be successfully used for the timely prediction of variable photosynthetic capacity and provide the fundamentals for monitoring and mapping plant functions from reflected information.

Partial least-squares (PLS) regression is a popular method for modeling chemical constituents from spectroscopic data and has been widely applied to retrieve leaf chemical components via hyperspectral remote sensing. However, one persistent challenge for applying the PLS regression is the selection of informative spectral bands among the vast array of acquired spectra. No consensus has been reached yet on how to select informative bands regardless of many techniques being proposed. In this paper, we have composited four individual data sets containing a total of 598 leaf samples from various species to evaluate four different band elimination/selection methods. Results revealed that the stepwise-PLS approach was optimal to estimate leaf chlorophyll content even under different spectral resolutions, from which informative bands were identified. Informative bands, in general, include bands inside the near-infrared (NIR), and in addition, one within the blue range and one within the red range. With such combinations, the PLS regression models meet the requirement for accurate leaf chlorophyll estimation. For most PLS regression models, their accuracies decreased with the reduction of spectral resolution, but the stepwise-PLS approach could consistently estimate the chlorophyll content at different spectral resolutions (with R2 ≥ 0.77 for resolutions <; 20 nm). The findings, hence, provide valuable insights for selecting informative spectral bands for PLS analysis and lay a strong foundation for retrieving foliar biochemical content using hyperspectral remote sensing data.