Low Resource Neural Machine Translation: Assamese to/from Other Indo-Aryan (Indic) Languages

Abstract

Machine translation (MT) systems have been built using numerous different techniques for bridging the language barriers. These techniques are broadly categorized into approaches like Statistical Machine Translation (SMT) and Neural Machine Translation (NMT). End-to-end NMT systems significantly outperform SMT in translation quality on many language pairs, especially those with the adequate parallel corpus. We report comparative experiments on baseline MT systems for Assamese to other Indo-Aryan languages (in both translation directions) using the traditional Phrase-Based SMT as well as some more successful NMT architectures, namely basic sequence-to-sequence model with attention, Transformer, and finetuned Transformer. The results are evaluated using the most prominent and popular standard automatic metric BLEU (BiLingual Evaluation Understudy), as well as other well-known metrics for exploring the performance of different baseline MT systems, since this is the first such work involving Assamese. The evaluation scores are compared for SMT and NMT models for the effectiveness of bi-directional language pairs involving Assamese and other Indo-Aryan languages (Bangla, Gujarati, Hindi, Marathi, Odia, Sinhalese, and Urdu). The highest BLEU scores obtained are for Assamese to Sinhalese for SMT (35.63) and the Assamese to Bangla for NMT systems (seq2seq is 50.92, Transformer is 50.01, and finetuned Transformer is 50.19). We also try to relate the results with the language characteristics, distances, family trees, domains, data sizes, and sentence lengths. We find that the effect of the domain is the most important factor affecting the results for the given data domains and sizes. We compare our results with the only existing MT system for Assamese (Bing Translator) and also with pairs involving Hindi.

English-Chinese Machine Translation Based on Transfer Learning and Chinese-English Corpus

“…However, with standard definition, interpretation can be completed in 50,000 steps, as shown in Figure 4. Due to the influence of parameter initialization before machine translation model training, the parameters of largescale Chinese-English translation model belonging to the same translation task are introduced into the initialization of low-resource Chinese-English and Tibetan-Chinese translation models so that the model has a certain parameter basis before training, so its learning rate will be improved during retraining [24,25]. In this document, the encoder and decoder parameters of the Chinese-English translation model are initialized together with the parameters of the Chinese encoder of the Sino-English model and the decoder of the English-Chinese model.…”

Findings of the WMT 2023 Shared Task on Low-Resource Indic Language Translation

“…Assamese exhibits a subject-object-verb (SOV) word order, in contrast to the subject-verb-object (SVO) word order found in English. Additionally, it is characterized as an agglutinative language, as discussed by Sarma et al (2017) and Baruah et al (2021), signifying its propensity to incorporate suffixes and prefixes into words to convey diverse grammatical meanings. This intricacy poses a notable challenge for machine translation systems, as they must accurately analyze and generate these intricate word forms.…”

NITS-CNLP Low-Resource Neural Machine Translation Systems of English-Manipuri Language Pair

“…MT researchers have introduced several approaches to overcome this bottleneck such as data augmentation using back-translation (Sennrich et al, 2016a), multilingual approach (Singh and Singh, 2022a), semi-supervised approach (Cheng et al, 2016; Singh andSingh, 2022b) and exploiting cues from multiple modalities (Gain et al, 2021; Meetei et al, 2023. There are also reports of a comparative study of MT systems on the low resource machine translation focusing on Indian languages such as Assamese (Baruah et al, 2021) and Mizo (Devi et al, 2022; Thangkhanhau andHussain, 2023).…”