Dialog systems for Vietnamese

Building a robust dialog system for an under-resourced language such as Vietnamese is a challenging task. This page gathers relevant information and resources for this task.

1. Speech recognition

Papers:

• The Effect of Tone Modeling in Vietnamese LVCSR System (Quoc Bao Nguyen et al. 216)

• A non-expert Kaldi recipe for Vietnamese Speech Recognition System (Hieu-Thi Luong and Hai-Quan Vu, WLSI/OIAF4HLT 2016)

• Automatic speech recognition of Vietnamese (Nguyen Thien Chuong, PhD thesis 2014)

• Vietnamese Large Vocabulary Continuous Speech Recognition (Ngoc Thang Vu and Tanja Schultz, ASRU 2009)

• A novel approach in continuous speech recognition for Vietnamese, an isolating tonal language (Nguyen Hong Quang et al. SLT 2008)

Corpora:

• Vivos (15 hours of read speech)

• Globalphone Vietnamese (22.5 hours of read speech from 15 Vietnamese online newspapers, not free)

• Vietnamese Speech Recognition Corpus- (Mobile)- 144 Speakers (76.6 hours)

• Vietnamese Speech Recognition Corpus-(In-Car)-300 Speakers (305 hours)

Tools: Kaldi, CMU Sphinx, OpenEars, HTK

2. Language understanding

• Identifying User Intents in Vietnamese Spoken Language Commands and Its Application in Smart Mobile Voice Interaction (Lan Ngo et al. ACIIDS 2016)

3. Dialog manager

4. Language generation

Not much work found for Vietnamese language generation. Here is some work in the other domains (machine translation)

• Generation of Vietnamese for French-Vietnamese and English-Vietnamese Machine Translation (Doan Nguyen Hai, EWNLG 2001)

5. Speech synthesis

Papers:

• HMM-based Vietnamese speech synthesis (Quoc Son Trinh, ICIS 2015)

• An HMM-based Vietnamese speech synthesis system (Thang Tat Vu et al. 2009)

Online services:

• Responsive Voice

6. Other NLP related tasks

Word segmentation

Papers:

• Vietnamese word segmentation with CRFs and SVMs: an investigation (Cam-Tu Nguyen et al. 2006)

• Vietnamese word segmentation (Dinh Dien et al. AFNLP 2001)

Part of Speech tagging

• An empirical study of maximum entropy approach for part-of-speech tagging of Vietnamese texts (Phuong Le-Hong et al., TALN 2010)

Name entity recognition

• Vietnamese Named Entity Recognition using Token Regular Expressions and Bidirectional Inference (Phuong Le-Hong, arxiv 2016)

Software and tools:

• Vitk