Cross-lingual transfer learning for spoken language understanding

2019/04/03 by Do, Quynh Ngoc Thi, Gaspers, Judith
#Computation and Language (cs.CL) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.1904.01825

Abstract

Typically, spoken language understanding (SLU) models are trained on annotated data which are costly to gather. Aiming to reduce data needs for bootstrapping a SLU system for a new language, we present a simple but effective weight transfer approach using data from another language. The approach is evaluated with our promising multi-task SLU framework developed towards different languages. We evaluate our approach on the ATIS and a real-world SLU dataset, showing that i) our monolingual models outperform the state-of-the-art, ii) we can reduce data amounts needed for bootstrapping a SLU system for a new language greatly, and iii) while multitask training improves over separate training, different weight transfer settings may work best for different SLU modules.

Related

• Large-scale Transfer Learning for Low-resource Spoken Language Understanding
• Evaluating Cross-Lingual Transfer Learning Approaches in Multilingual\n Conversational Agent Models
• Analysis of Multi-Source Language Training in Cross-Lingual Transfer
• Measuring Cross-lingual Transfer in Bytes
• From Masked Language Modeling to Translation: Non-English Auxiliary Tasks Improve Zero-shot Spoken Language Understanding