In this paper, we introduce our work of building a Streaming Multilingual Speech Model (SM2), which can transcribe or translate multiple spoken languages into texts of the target language. The backbone of SM2 is Transformer Transducer, which has high streaming capability. Instead of human labeled speech translation (ST) data, SM2 models are trained using weakly supervised data generated by converting the transcriptions in speech recognition corpora with a machine translation service. With 351 thousand hours of anonymized speech training data from 25 languages, SM2 models achieve comparable or even better ST quality than some recent popular large-scale non-streaming speech models. More importantly, we show that SM2 has the truly zero-shot capability when expanding to new target languages, yielding high quality ST results for {source-speech, target-text} pairs that are not seen during training.

Gender inclusivity in language technologies has become a prominent research topic. In this study, we explore gender-neutral translation (GNT) as a form of gender inclusivity and a goal to be achieved by machine translation (MT) models, which have been found to perpetuate gender bias and discrimination. Specifically, we focus on translation from English into Italian, a language pair representative of salient gender-related linguistic transfer problems. To define GNT, we review a selection of relevant institutional guidelines for gender-inclusive language, discuss its scenarios of use, and examine the technical challenges of performing GNT in MT, concluding with a discussion of potential solutions to encourage advancements toward greater inclusivity in MT.

Language technology carries a growing responsibility in a society that is increasingly dominated by digital communication channels. Machine translation (MT) plays a decisive role in cross-lingual information access and will continue to grow as a crucial component in our natural language processing (NLP) toolbox, enabling inclusiveness and equity among people with different cultural and linguistic backgrounds. All the major IT companies recognize the importance of MT and push significant efforts into the development of internal translation solutions with slogans like "no language left behind"

The ability to extract high-quality translation dictionaries from monolingual word embedding spaces depends critically on the geometric similarity of the spaces -- their degree of "isomorphism." We address the root-cause of faulty cross-lingual mapping: that word embedding training resulted in the underlying spaces being non-isomorphic. We incorporate global measures of isomorphism directly into the Skip-gram loss function, successfully increasing the relative isomorphism of trained word embedding spaces and improving their ability to be mapped to a shared cross-lingual space. The result is improved bilingual lexicon induction in general data conditions, under domain mismatch, and with training algorithm dissimilarities. We release IsoVec at https://github.com/kellymarchisio/isovec.

The major hindrance in the study of earlier scientific literature is the availability of Latin translations into modern languages. This is particular true for the works of Euler who authored about 850 manuscripts and wrote a thousand letters and received back almost two thousand more. The translation of many of these manuscripts, books and letters have been published in various sources over the last two centuries, but many more have not yet appeared. Fortunately, nowadays, the artificial intelligence AI translation can be used to circumvent the challenges of translating such substantial number of texts. To validate this tool, benchmark tests have been performed to compare the performance of two popular AI translating algorithms, namely Google Translate and ChatGPT. Since it was found that ChatGPT performed better on these tests, this translating support was then used on an excerpt of a 1739 letter from Johann Bernoulli to Euler, where he notifies that he was sending to Euler the first part of his manuscript Hydraulica. The findings highlight ChatGPT as a valuable translation tool, catering not only to general Latin practitioners but also proving beneficial for specialized Latin translators.

Simultaneous speech translation is an essential communication task difficult for humans whereby a translation is generated concurrently with oncoming speech inputs. For such a streaming task, transformers using block processing to break an input sequence into segments have achieved state-of-the-art performance at a reduced cost. Current methods to allow information to propagate across segments, including left context and memory banks, have faltered as they are both insufficient representations and unnecessarily expensive to compute. In this paper, we propose an Implicit Memory Transformer that implicitly retains memory through a new left context method, removing the need to explicitly represent memory with memory banks. We generate the left context from the attention output of the previous segment and include it in the keys and values of the current segment's attention calculation. Experiments on the MuST-C dataset show that the Implicit Memory Transformer provides a substantial speedup on the encoder forward pass with nearly identical translation quality when compared with the state-of-the-art approach that employs both left context and memory banks.

Transformer models using segment-based processing have been an effective architecture for simultaneous speech translation. However, such models create a context mismatch between training and inference environments, hindering potential translation accuracy. We solve this issue by proposing Shiftable Context, a simple yet effective scheme to ensure that consistent segment and context sizes are maintained throughout training and inference, even with the presence of partially filled segments due to the streaming nature of simultaneous translation. Shiftable Context is also broadly applicable to segment-based transformers for streaming tasks. Our experiments on the English-German, English-French, and English-Spanish language pairs from the MUST-C dataset demonstrate that when applied to the Augmented Memory Transformer, a state-of-the-art model for simultaneous speech translation, the proposed scheme achieves an average increase of 2.09, 1.83, and 1.95 BLEU scores across each wait-k value for the three language pairs, respectively, with a minimal impact on computation-aware Average Lagging.

The text inputs are often context to understand the world. From the simple and sufficient for translation tasks (Wu perspective of NMT, it is also much needed to et al., 2021). Take the widely used Multi30K as make use of such information to approach humanlevel an example. Multi30K consists of only 30K image translation abilities. To facilitate Multimodal captions, while typical text translation systems are Machine Translation (MMT) research, a number often trained with several million sentence pairs. of datasets have been proposed including imageguided We argue that studying the effects of visual contexts translation datasets (Elliott et al., 2016; in machine translation requires a large-scale Gella et al., 2019; Wang et al., 2022) and videoguided and diverse data set for training and a real-world translation datasets (Sanabria et al., 2018; and complex benchmark for testing.

The rapid growth of machine translation (MT) systems has necessitated comprehensive studies to meta-evaluate evaluation metrics being used, which enables a better selection of metrics that best reflect MT quality. Unfortunately, most of the research focuses on high-resource languages, mainly English, the observations for which may not always apply to other languages. Indian languages, having over a billion speakers, are linguistically different from English, and to date, there has not been a systematic study of evaluating MT systems from English into Indian languages. In this paper, we fill this gap by creating an MQM dataset consisting of 7000 fine-grained annotations, spanning 5 Indian languages and 7 MT systems, and use it to establish correlations between annotator scores and scores obtained using existing automatic metrics. Our results show that pre-trained metrics, such as COMET, have the highest correlations with annotator scores. Additionally, we find that the metrics do not adequately capture fluency-based errors in Indian languages, and there is a need to develop metrics focused on Indian languages. We hope that our dataset and analysis will help promote further research in this area.

Transformers belong to a class of machine learning models that use self-attention or the scaled dot-product operation as their primary learning mechanism. Transformers were initially proposed for neural machine translation - one of the most challenging natural language processing (NLP) tasks [1]. Recently, Transformers have been successfully employed to tackle various problems in machine learning and achieve state-of-the-art performance [2]. Apart from classical NLP tasks, examples from other areas include image classification [3], object detection and segmentation [4], image and language generation [5], sequential decision-making in reinforcement learning [6], multi-modal (text, speech, and image) data processing [7], and analysis of tabular and time-series data [8]. This tutorial paper focuses on time-series analysis using Transformers. Time-series data consist of ordered samples, observations, or features recorded sequentially over time. Time-series datasets often arise naturally in many real-world applications where data is recorded over a fixed sampling interval. Examples include stock prices, digitized speech signals, traffic measurements, sensor data for weather patterns, biomedical measurements, and various kinds of population data recorded over time.