Vietnamese Profanity Filter API

Detect and mask Vietnamese (and common English) profanity, insults and slurs, including unaccented spellings, teencode, l33t and separators like đ.ị.t.

GETPOSThttps://apibox.io.vn/v1/profanity-vi

Text▶ Try it online

The profanity filter finds offensive words in Vietnamese comments, chats and reviews and returns a masked version. Its lexicon covers sexual and vulgar words, insults and teencode abbreviations (đm, vcl, dmm, clgt…) plus common English swear words and slurs, each with a category (sexual, vulgar, insult, slur) and a severity (mild, moderate, severe).

It is built for the usual tricks: missing accents where the result is still unambiguous (dit me), repeated letters (đụuuu), l33t digits and symbols (sh1t, f@ck), and separators inside words (đ.ị.t, l-ồ-n). Words that are only offensive with their accents (lồn, cặc, buồi) are matched with accents only, so “lon bia”, “các bạn” and “buổi sáng” are never flagged; ambiguous words (cc, đù, cave) are only checked with strict=true.

The response gives the overall severity, a score, each match with its position and the censored text (mask character, keep_first=true to keep the first letter). Add your own words with extra_words and whitelist with allow.

Parameters

Name	In	Type	Required	Default
`text`	body\|query	string	yes		Text to check (max 100,000 characters).
`mask`	body\|query	string	no	`*`	Masking character.
`keep_first`	body\|query	boolean	no	`false`	Keep the first letter of masked words.
`strict`	body\|query	boolean	no	`false`	Also flag ambiguous words (cc, đù, cave…).
`extra_words`	body\|query	string	no		Comma-separated words to add.
`allow`	body\|query	string	no		Comma-separated words to ignore.

Example

curl -X POST "https://apibox.io.vn/v1/profanity-vi" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text":"Đm, cái này vcl thật","keep_first":true}'

const res = await fetch('https://apibox.io.vn/v1/profanity-vi', {
  method: 'POST',
  headers: {
    Authorization: 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    "text": "Đm, cái này vcl thật",
    "keep_first": true
  }),
});
const data = await res.json();
console.log(data);

import requests

res = requests.post(
    "https://apibox.io.vn/v1/profanity-vi",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={"text": "Đm, cái này vcl thật", "keep_first": True},
    timeout=30,
)
print(res.json())

<?php
$ch = curl_init('https://apibox.io.vn/v1/profanity-vi');
curl_setopt_array($ch, [
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_CUSTOMREQUEST => 'POST',
    CURLOPT_HTTPHEADER => ['Authorization: Bearer YOUR_API_KEY', 'Content-Type: application/json'],
    CURLOPT_POSTFIELDS => '{"text":"Đm, cái này vcl thật","keep_first":true}',
]);
$response = curl_exec($ch);
print_r(json_decode($response, true));

Response

{
  "profane": true,
  "severity": "moderate",
  "score": 60,
  "count": 2,
  "categories": [
    "vulgar"
  ],
  "matches": [
    {
      "text": "Đm",
      "start": 0,
      "end": 2,
      "word": "dm",
      "category": "vulgar",
      "severity": "moderate"
    },
    {
      "text": "vcl",
      "start": 12,
      "end": 15,
      "word": "vcl",
      "category": "vulgar",
      "severity": "moderate"
    }
  ],
  "censored": "Đ*, cái này v** thật",
  "strict": false
}

Use cases

Moderate comments, reviews and chat messages
Filter usernames and display names at sign-up
Flag messages for human review in support tools
Keep bot and game chats family-friendly

FAQ

Will it flag normal words without accents?

No. Words like lồn / cặc / buồi only count with their accents, because lon, cac, buoi are ordinary words without them.

Can I tune the list?

Yes: extra_words adds comma-separated words or phrases, allow removes entries, strict=true enables ambiguous ones.

Is the text stored?

No, it is processed in memory only.

Vietnamese Profanity Filter API

Parameters

Example

Response

Use cases

FAQ

Related APIs

Text Diff

Slugify

Text Statistics