Search Constraints
Number of results to display per page
Results for:
Tags sim
Linguistic
Remove constraint Tags sim: Linguistic
1 - 2 of 2
Search Results
-
Video
People are now regularly interacting with voice assistants (VAs), which are conversational agents that allow users to use spoken language to interface with a machine to complete tasks. The huge adoption and daily use of VAs by millions of people - and its increasing use for financial, healthcare, and educational applications - raises important questions about the linguistic and social factors that affect spoken language interactions with machines.
We are exploring issues of linguistic and social biases that impact speech communication in human-computer interaction - particularly during cross-language transfer, learning, or adaptation of some kind. In this talk, I will present two case studies illustrating some of our most recent work in this area. The first study looks at a case of cross-language ASR transfer. We find systematic linguistic and phonetic disparities in language transfer by machines trained on a source language to speech recognition of a novel target, low-resource language. The second study looks at a case of social bias in word learning by humans using voice-enabled apps. We find the word learning is inhibited when there are mismatching social cues presented by the voice and the linguistic information.
Together, along with highlights from other ongoing work in my lab, the aim of this talk is to underscore that human-computer linguistic communication is a rich testing ground for investigating issues in speech and language variation. Examining linguistic variation during HCI can enrich and elaborate linguistic theory, as well as present opportunities for linguists to provide insights for improving both the function and fairness of these technologies.
Event date: 25/03/2025
Speaker: Professor. Georgia ZELLOU (University of California, Davis)
Hosted by: Faculty of Humanities
- Subjects:
- Communication and Language and Languages
- Keywords:
- Linguistics English language -- Variation Speech processing systems English language -- Spoken English Human-computer interaction
- Resource Type:
- Video
-
Others
This database is constructed on the basis of two earlier databases developed by the Research Centre for the Humanities Computing (formerly the Humanities Computing Programme). Since the appearance of the Chinese Syllabary Pronounced according to the Dialect of Canton in 1996 and the Chinese Talking Syllabary of the Cantonese Dialect: An Electronic Repository in 1998, we have been receiving notes of gratitude from users from all over the Internet. Out of the many suggestions they made, the crucial one was the expansion of our databases from a pure syllabary structure into one which covers semantic information of the characters. In response to this suggestion of our users, in particular their concern for the semantic disambiguation of Chinese polyphonic characters, a database carrying the current title was planned. Being functionally versatile and user-friendly like its two predecessors, the current new database excels further in the following respects: This fully revised and expanded database covering the complete Big5 Chinese character set is now the most comprehensive syllabary of the Cantonese dialect on the Internet. It covers in the first place the syllabric data of four major works, namely, 1) S. L. Wong's A Chinese Syllabary Pronounced according to the Dialect of Canton, 2) Li Chomin's Lishi Zhongwen Zidian, 3) Zhou Wuji and Rao Bingcai's Guangzhou Hua Biaozunyin Zihui and 4) Richard Ho and Chu Kwok-fan's Yuehyin Zhengdu Zihui. To make up what is still missing, linguistic information of nine other major works are consulted. To take into account the linguistic reality of the Hong Kong society, vernacular pronunciation data provided by the Linguistic Society of Hong Kong are also included. Besides pronunciations, typical word-forms or vocabularies are provided for every character in this database. These word-forms are grouped with respect to the proper pronunciation(s) of the respective head characters so that users can disambiguate polyphonic characters that are phonologically ambiguous. In cases where common vocabularies are not readily available, brief remarks or explanations will be given. It supports up to seven transciption (romanization) schemes of the Cantonese dialect. Users can switch from one scheme to the other wherever necessary. When a certain head character is being featured, basic information such as pronunciation(s), homophones, vocabularies etc. are tabulated. In addition to these, further lexical information related to that particular character will also be provided for easy reference, as, for instances, internal codes (Big5 and Unicode), Cangjie input code, radical belonging , number of strokes, basic English translation, pagination of important references and hyperlinks pointing to various online resources. We would like to extend our sincere thanks to Ms. Ginny Chan, former instructor of Yale-China Chinese Language Center, CUHK, for her courtesy in demonstrating 1,900 unique Cantonese pronunciations on a volunteer basis.
- Course related:
- CBS4901 Contrastive Analysis of Chinese and English, CBS 3407 Chinese Academic Writing in Language and Speech Science, CBS532 Description of Chinese I: Words and Sentences, and CBS514 Introduction to Cantonese studies
- Subjects:
- Chinese Language
- Keywords:
- Chinese language -- Dialects Dictionaries Cantonese dialects -- Pronunciation
- Resource Type:
- Others