Open AccessSmåprat: DialoGPT for Natural Language Generation of Swedish Dialogue by Transfer Learning
Author(s) -
Tosin Adewumi,
Nosheen Abid,
Maryam Pahlavan,
Rickard Brännvall,
Sana Sabah Sabry,
Foteini Liwicki,
Marcus Liwicki
Publication year - 2022
Publication title -
proceedings of the northern lights deep learning workshop
Language(s) - English
Resource type - Journals
ISSN - 2703-6928
DOI - 10.7557/18.6231
Subject(s) - perplexity , chatbot , computer science , transfer of learning , language model , natural language processing , transformer , natural language , artificial intelligence , open domain , language understanding , metric (unit) , question answering , engineering , operations management , voltage , electrical engineering
Building open-domain conversational systems (or chatbots) that produce convincing responses is a recognized challenge. Recent state-of-the-art (SoTA) transformer-based models for the generation of natural language dialogue have demonstrated impressive performance in simulating human-like, single-turn conversations in English. This work investigates, by an empirical study, the potential for transfer learning of such models to Swedish language. DialoGPT, an English language pre-trained model, is adapted by training on three different Swedish language conversational datasets obtained from publicly available sources. Perplexity score (an automated intrinsic language model metric) and surveys by human evaluation were used to assess the performances of the fine-tuned models, with results that indicate that the capacity for transfer learning can be exploited with considerable success. Human evaluators asked to score the simulated dialogue judged over 57% of the chatbot responses to be human-like for the model trained on the largest (Swedish) dataset. We provide the demos and model checkpoints of our English and Swedish chatbots on the HuggingFace platform for public use.