The Berber, or Amazigh language family is a low-resource North African vernacular language spoken by the indigenous Berber ethnic group. It has its own unique alphabet called Tifinagh used across Berber communities in Morocco, Algeria, and others. The Afroasiatic language Berber is spoken by 14 million people, yet lacks adequate representation in education, research, web applications etc. For instance, there is no option of translation to or from Amazigh / Berber on Google Translate, which hosts over 100 languages today. Consequently, we do not find specialized educational apps, L2 (2nd language learner) acquisition, automated language translation, and remote-access facilities enabled in Berber. Motivated by this background, we propose a supervised approach called DaToBS for Detection and Transcription of Berber Signs. The DaToBS approach entails the automatic recognition and transcription of Tifinagh characters from signs in photographs of natural environments. This is achieved by self-creating a corpus of 1862 pre-processed character images; curating the corpus with human-guided annotation; and feeding it into an OCR model via the deployment of CNN for deep learning based on computer vision models. We deploy computer vision modeling (rather than language models) because there are pictorial symbols in this alphabet, this deployment being a novel aspect of our work. The DaToBS experimentation and analyses yield over 92 percent accuracy in our research. To the best of our knowledge, ours is among the first few works in the automated transcription of Berber signs from roadside images with deep learning, yielding high accuracy. This can pave the way for developing pedagogical applications in the Berber language, thereby addressing an important goal of outreach to underrepresented communities via AI in education.

The remarkable success of transformers in the field of natural language processing has sparked the interest of the speech-processing community, leading to an exploration of their potential for modeling long-range dependencies within speech sequences. Recently, transformers have gained prominence across various speech-related domains, including automatic speech recognition, speech synthesis, speech translation, speech para-linguistics, speech enhancement, spoken dialogue systems, and numerous multimodal applications. In this paper, we present a comprehensive survey that aims to bridge research studies from diverse subfields within speech technology. By consolidating findings from across the speech technology landscape, we provide a valuable resource for researchers interested in harnessing the power of transformers to advance the field. We identify the challenges encountered by transformers in speech processing while also offering insights into potential solutions to address these issues.

Sign Language Translation (SLT) is a promising technology to bridge the communication gap between the deaf and the hearing people. Recently, researchers have adopted Neural Machine Translation (NMT) methods, which usually require large-scale corpus for training, to achieve SLT. However, the publicly available SLT corpus is very limited, which causes the collapse of the token representations and the inaccuracy of the generated tokens. To alleviate this issue, we propose ConSLT, a novel token-level \textbf{Con}trastive learning framework for \textbf{S}ign \textbf{L}anguage \textbf{T}ranslation , which learns effective token representations by incorporating token-level contrastive learning into the SLT decoding process. Concretely, ConSLT treats each token and its counterpart generated by different dropout masks as positive pairs during decoding, and then randomly samples $K$ tokens in the vocabulary that are not in the current sentence to construct negative examples. We conduct comprehensive experiments on two benchmarks (PHOENIX14T and CSL-Daily) for both end-to-end and cascaded settings. The experimental results demonstrate that ConSLT can achieve better translation quality than the strong baselines.

Simultaneous machine translation, which aims at a real-time translation, is useful in many live scenarios but very challenging due to the trade-off between accuracy and latency. To achieve the balance for both, the model needs to wait for appropriate streaming text (READ policy) and then generates its translation (WRITE policy). However, WRITE policies of previous work either are specific to the method itself due to the end-to-end training or suffer from the input mismatch between training and decoding for the non-end-to-end training. Therefore, it is essential to learn a generic and better WRITE policy for simultaneous machine translation. Inspired by strategies utilized by human interpreters and "wait" policies, we propose a novel adaptive prefix-to-prefix training policy called LEAPT, which allows our machine translation model to learn how to translate source sentence prefixes and make use of the future context. Experiments show that our proposed methods greatly outperform competitive baselines and achieve promising results.

To make robots accessible to a broad audience, it is critical to endow them with the ability to take universal modes of communication, like commands given in natural language, and extract a concrete desired task specification, defined using a formal language like linear temporal logic (LTL). In this paper, we present a learning-based approach for translating from natural language commands to LTL specifications with very limited human-labeled training data. This is in stark contrast to existing natural-language to LTL translators, which require large human-labeled datasets, often in the form of labeled pairs of LTL formulas and natural language commands, to train the translator. To reduce reliance on human data, our approach generates a large synthetic training dataset through algorithmic generation of LTL formulas, conversion to structured English, and then exploiting the paraphrasing capabilities of modern large language models (LLMs) to synthesize a diverse corpus of natural language commands corresponding to the LTL formulas. We use this generated data to finetune an LLM and apply a constrained decoding procedure at inference time to ensure the returned LTL formula is syntactically correct. We evaluate our approach on three existing LTL/natural language datasets and show that we can translate natural language commands at 75\% accuracy with far less human data ($\le$12 annotations). Moreover, when training on large human-annotated datasets, our method achieves higher test accuracy (95\% on average) than prior work. Finally, we show the translated formulas can be used to plan long-horizon, multi-stage tasks on a 12D quadrotor.

Coreference resolution aims to identify words and phrases which refer to same entity in a text, a core task in natural language processing. In this paper, we extend this task to resolving coreferences in long-form narrations of visual scenes. First we introduce a new dataset with annotated coreference chains and their bounding boxes, as most existing image-text datasets only contain short sentences without coreferring expressions or labeled chains. We propose a new technique that learns to identify coreference chains using weak supervision, only from image-text pairs and a regularization using prior linguistic knowledge. Our model yields large performance gains over several strong baselines in resolving coreferences. We also show that coreference resolution helps improving grounding narratives in images.

As it is cumbersome and expensive to acquire a huge amount of data for training neural dialog models, data augmentation is proposed to effectively utilize existing training samples. However, current data augmentation techniques on the dialog generation task mostly augment all cases in the training dataset without considering the intrinsic attributes between different cases. We argue that not all cases are beneficial for augmentation task, and the cases suitable for augmentation should obey the following two attributes: (1) low-quality (the dialog model cannot generate a high-quality response for the case), (2) representative (the case should represent the property of the whole dataset). Herein, we explore this idea by proposing a Selective Data Augmentation framework (SDA) for the response generation task. SDA employs a dual adversarial network to select the lowest quality and most representative data points for augmentation in one stage. Extensive experiments conducted on two publicly available datasets, i.e., DailyDialog and OpenSubtitles, show that our framework can improve the response generation performance with respect to various metrics.

Vulnerability to lexical perturbation is a critical weakness of automatic evaluation metrics for image captioning. This paper proposes Perturbation Robust Multi-Lingual CLIPScore(PR-MCS), which exhibits robustness to such perturbations, as a novel reference-free image captioning metric applicable to multiple languages. To achieve perturbation robustness, we fine-tune the text encoder of CLIP with our language-agnostic method to distinguish the perturbed text from the original text. To verify the robustness of PR-MCS, we introduce a new fine-grained evaluation dataset consisting of detailed captions, critical objects, and the relationships between the objects for 3, 000 images in five languages. In our experiments, PR-MCS significantly outperforms baseline metrics in capturing lexical noise of all various perturbation types in all five languages, proving that PR-MCS is highly robust to lexical perturbations.

Transformers are a family of recently introduced Deep Learning (DL) models which leverage the mechanism of dot-product attention to map a sequence of tokens of arbitrary length into a new set of tokens. Thanks to their outstanding performance in a variety of tasks, transformers are nowadays ubiquitous in state-of-the-art techniques that gain any benefit from modeling long-term interactions between elements of a sequence. Another important advantage of transformers is the ability to process sequences of arbitrary length in a single forward pass without incurring the limitations of recurrent approaches: no other standard Machine Learning (ML) or DL methods in the literature have shown this great adaptability so far. In the domain of Natural Language Processing (NLP) transformers are pervasive in any sort of task, such as Machine Translation [1-4], text classification, document retrieval, document summarization and several others more. More recently, researchers started to focus on exploiting the benefits of the self-attention mechanism for computer vision tasks [5-7], either standalone or applied downstream to a convolutional backbone and even to multimodal problems where the language and visual input needs to be correlated.

Non-autoregressive neural machine translation (NAT) models are proposed to accelerate the inference process while maintaining relatively high performance. However, existing NAT models are difficult to achieve the desired efficiency-quality trade-off. For one thing, fully NAT models with efficient inference perform inferior to their autoregressive counterparts. For another, iterative NAT models can, though, achieve comparable performance while diminishing the advantage of speed. In this paper, we propose RenewNAT, a flexible framework with high efficiency and effectiveness, to incorporate the merits of fully and iterative NAT models. RenewNAT first generates the potential translation results and then renews them in a single pass. It can achieve significant performance improvements at the same expense as traditional NAT models (without introducing additional model parameters and decoding latency). Experimental results on various translation benchmarks (e.g., \textbf{4} WMT) show that our framework consistently improves the performance of strong fully NAT methods (e.g., GLAT and DSLP) without additional speed overhead.