Neural machine translation (NMT) typically relies on a substantial number of bilingual parallel corpora for effective training. Mongolian, as a low-resource language, has relatively few parallel corpora, resulting in poor translation performance. Data augmentation (DA) is a practical and promising method to solve problems related to data sparsity and single semantic structure by expanding the size and structure of available data. In order to address the issues of data sparsity and semantic inconsistency in Mongolian–Chinese NMT processes, this paper proposes a new semantic-context DA method. This method adds an additional semantic encoder based on the original translation model, which utilizes both source and target sentences to generate different semantic vectors to enhance each training instance. The results show that this method significantly improves the quality of Mongolian–Chinese NMT tasks, with an increase of approximately 2.5 BLEU values compared to the basic Transformer model. Compared to the basic model, this method can achieve the same translation results with about half of the data, greatly improving translation efficiency.

To address the challenges of slow synthesis speed, unstable quality, limited emotional expressiveness, and the lack of controllable speaking rate in Mongolian emotional speech synthesis, this paper proposes a speech Rate-Controllable Mongolian emotional speech synthesis model based on improved Tacotron2 (SRC-IT2). First, an end-to-end Mongolian speech synthesis module is constructed based on an improved Tacotron2 framework, incorporating the unique linguistic characteristics of the Mongolian script. The front-end processing is optimized accordingly, and a G2P-Seq2Seq model is employed to achieve accurate grapheme-to-phoneme conversion for Mongolian characters. Next, on top of the end-to-end synthesis framework, a joint text-audio emotion analysis module is integrated to effectively learn and represent emotional style features specific to Mongolian speech. Finally, a style encoder and speaking rate control variable are embedded into the acoustic modeling process, further enhancing Tacotron2’s ability to dynamically adjust the speaking rate during emotional speech generation. Experimental results demonstrate that the proposed model produces more natural-sounding speech with improved emotional expressiveness and enables effective real-time control over speaking rate in Mongolian emotional speech synthesis.

Serious information loss often occurs when the input data undergoes the layer-by-layer feature extraction process through deep neural networks. In small-target detection tasks, this problem becomes more serious. The existing theory holds that the information bottleneck leads to the decrease of an algorithm’s recognition rate. In order to improve the precision of small-target recognition, PGI-ViMamba is proposed in this paper to resist the information attenuation of a neural network model. The model mainly uses Multi-Level Attention-Gated Programmable Gradient Information (MLAG PGI) and SPD-Conv-VSS module. Specifically, the backbone uses an improved VSS module as a feature extractor and uses an auxiliary branch similar to YOLOv9. This neural network structure ensures that the neural network retains the feature information of the small target in the process of forward propagation while reducing the model parameters. Another advantage is that the MLAG PGI acts as a reversible branch, providing reliable gradient information. Experiments show the effectiveness of the algorithm. Compared with state-of-the-art (SOTA) models, the proposed method achieves improvements of 1.1% and 2.1% in small-target recognition precision on the VisDrone and DOTA-v1.5 datasets, respectively, with no significant decline in recall rates. Additionally, ablation experiments validate the effectiveness of the MLAG PGI and SPD-Conv-VSS modules.

In Mongolian Text Sentiment Analysis (MTSA), the scarcity of annotated sentiment datasets and the insufficient consideration of syntactic dependency and topological structural information pose significant challenges to accurately capturing semantics and effectively extracting emotional features. To address these issues, this paper proposes a Mongolian Text Sentiment Analysis model based on ConvBERT and Graph Attention Network (MTSA-CG). Firstly, the ConvBERT pre-trained model is employed to extract textual features under limited data conditions, aiming to mitigate the shortcomings caused by data scarcity. Concurrently, textual data are transformed into graph-structured data, integrating co-occurrence, dependency, and similarity information into a Graph Attention Network (GAT) to capture syntactic and structural cues, enabling a deeper understanding of semantic and emotional connotations for more precise sentiment classification. The proposed multi-graph fusion strategy employs a hierarchical attention mechanism that dynamically weights different graph types based on their semantic relevance, distinguishing it from conventional graph aggregation methods. Experimental results demonstrate that, in comparison with various advanced baseline models, the proposed method significantly enhances the accuracy of MTSA.