Dataset Viewer
The dataset viewer is not available for this dataset.
Unexpected token '<', "<html> <h"... is not valid JSON

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Hong Kong Cantonese Corpus (HKCanCor) - Liujgoj Romanization Version

註:2026-07-05更正:將黐住嘅英文單詞分開咗。

項目簡介 / Project Overview

本項目基於 Hong Kong Cantonese Corpus (HKCanCor) 香港粵語語料庫(原始項目開源於 fcbond/hkcancor,由 Luke Kang Kwong 教授創建)。

我對原始語料庫中 utf8 文件夾內的 58 個文本文件進行了處理,將原本的粵拼(Jyutping)轉寫為**溜歌羅馬字 (Liujgoj Romanization)**,並針對分詞與符號進行了優化調整,使其更適合用於粵語語音合成(TTS)、語音識別(ASR)或自然語言處理(NLP)相關任務。

This project is derived from the Hong Kong Cantonese Corpus (HKCanCor) (originally hosted at fcbond/hkcancor and created by Prof. Luke Kang Kwong). We have processed the 58 text files inside the utf8 folder, converting the original Jyutping into Liujgoj Romanization, along with word segmentation updates and symbol modifications to better suit Cantonese NLP and Speech tasks.


修改說明 / Modifications

根據 CC BY 4.0 協議要求,在此說明本版本對原始數據進行的修改: In accordance with the CC BY 4.0 license, here are the specific modifications made to the original dataset:

  1. **連接符號替換 (Word Connection)**:
    • 使用 -(連字號)代替原本的 0 來表示單詞內部的連接。
  2. **停頓符號替換 (Prosodic Pause)**:
    • 使用 /(前後帶空格的斜線)代替原本的 - 來表示語流中的停頓。
  3. **重新分詞 (Word Re-segmentation)**:
    • 修正並優化了部分單詞的分詞顆粒度。
    • 例 1(多字詞拆分)cinjkeixbaakgwaai 修正為 cinjkeix-baakgwaai
    • 例 2(單字詞分離)nijgo 修正為 nij go

數據格式 / Data Format

數據集採用「漢字文本」與「溜歌羅馬字」逐句對應的格式(每句包含說話者標籤 A:B:)。

Each dialogue turn consists of the original Cantonese text and its corresponding Liujgoj Romanization:

A:喂遲啲去唔去旅行啊?你老公有冇平機票啊?
A: Waix, cixdij heoi mx heoi leoiqhangx aa? Neiq louqgungj yauq mouq pengx geijpiu aa?
B:平機票要淡季先有得平𡃉喎。而家旺 / 。
B: Pengx geijpiu yiu daamhgwai sinj yauqdakj pengx gaa wo. Yixgaaj wongh / .

引用 / Citation

如果您在研究或項目中使用了本數據集,請務必按照原作者要求引用以下論文: If you use this dataset, please cite the original paper as required by the authors: K. K. Luke and May L.Y. Wong (2015) The Hong Kong Cantonese Corpus: Design and Uses Journal of Chinese Linguistics Monograph Series, 25, 312–333. (preprint, JSTOR)


許可證 / License

本項目數據繼承原始項目的 CC BY 4.0 (Creative Commons Attribution 4.0 International) 許可證。您可以自由分享和修改本數據,但必須給予原作者適當的署名(包含引用上述論文),並指出您所做的修改。

This dataset is released under the CC BY 4.0 (Creative Commons Attribution 4.0 International) license, inheriting the terms of the original HKCanCor project. You are free to share and adapt the material for any purpose, provided that you give appropriate credit, provide a link to the license, and indicate if changes were made.

Downloads last month
219