Skip to main content

Python implementation of the Chinese soundex project DimSim

Project description

DimSim - A Chinese Soundex Library (Python version)

DimSim is a library developed by the Scalable Knowledge Intelligence team at IBM Almaden Research Center as part of the SystemT project.

The PyPi project page can be found here. It was created in collaboration with IBM Center for Open-Source Data and AI Technologies (CODAIT).


We provide a phonetic algorithm for indexing Chinese characters by sound. The technical details can be found in the following paper:

Min Li, Marina Danilevsky, Sara Noeman and Yunyao Li. DIMSIM: An Accurate Chinese Phonetic Similarity Algorithm based on Learned High Dimensional Encoding. CoNLL 2018.

In this library, we provide a pre-trained model that can perform the following functions:

  • Given two Chinese phrases (of the same length), return the phonetic distance of the input phrases. Optionally you can feed in pinyin strings of Chinese phrases too.
  • Given a Chinese phrase, return its top-k similar (phoentically) Chinese phrases.

How to install


  • pypinyin: used for translating Chinese characters into their correponding pinyins.

There are two ways to install this library:

  • Install from PyPi
pip install dimsim
  • Download the source code by cloning this repo and compile it yourself.
git clone

cd DimSim/

pip install -e .

How to use

Once you have the package installed you can use it for the two functions as shown below.

  • Computing phonetic distance of two Chinese phrases. The optional argument pinyin (False by default) can be used to provide a pinyin string list directly. See example usage below.
import dimsim

dist = dimsim.get_distance("大侠","大虾")

dist = dimsim.get_distance("大侠","大人")

dist = dimsim.get_distance(['da4','xia2'],['da4','xia1']], pinyin=True)

dist = dimsim.get_distance(['da4','xia2'],['da4','ren2']], pinyin=True)

  • Return top-k phonetically similar phrases of a given Chinese phrase. Two parameters:
  • mode controls the character type of the returned Chinese phrases, where 'simplified' represents simplified Chinese and 'traditional' represents traditional Chinese.
  • theta controls the size of search space for the candidate phrases.
import dimsim

candidates = dimsim.get_candidates("大侠", mode="simplified", theta=1)
['打下', '大虾', '大侠']

candidates = dimsim.get_candidates("粉丝", mode="traditinoal", theta=1)
['門市', '分時', '焚屍', '粉飾', '粉絲']


Please cite the library by referencing the published paper:

  author = 	{Li, Min and Danilevsky, Marina and Noeman, Sara and Li, Yunyao},
  title = 	{{DIMSIM:} An Accurate Chinese Phonetic Similarity Algorithm Based on Learned High Dimensional Encoding},
  booktitle = 	{Proceedings of the 22nd Conference on Computational Natural Language Learning},
  year = 	{2018},
  publisher = 	{Association for Computational Linguistics},
  pages = 	{444-453},
  location = 	{Brussels, Belgium},
  url = 	{}

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for dimsim, version 0.2.2
Filename, size File type Python version Upload date Hashes
Filename, size dimsim-0.2.2-py3-none-any.whl (2.4 MB) File type Wheel Python version py3 Upload date Hashes View
Filename, size dimsim-0.2.2.tar.gz (2.4 MB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page