API Documentation

SentiVox's API provide 10,000 free tokens daily providing simple, fast, mulitlingual and accurate package to perform sentiment analysis on the text along with a word cloud.

API URL

https://senti-vox-api.onrender.com

Supported Languages

EnglishHindiMarathiTamil

API Endpoints

/GET - Returns the description of the API
/cf/singlePOST - Context free analysis of a single comment
/cf/batchPOST - Context free analysis of multiple comments
/healthGET - Health check
/docsGET - Documentations
/langsGET - Supported languages
/tokenizePOST- Tokenizes the text

Utility Endpoints

Tokenization

POST /tokenize

Divides/tokenizes the text into a list of tokens

{
  "api-key": "string - Your API key",
  "comment": "text to tokenize"
}

Token Count

POST /tokenCount

Returns the remaining tokens associated with your API Key

{
  "api-key": "string - Your API key"
}

Languages

GET /langs

Returns the list of all supported languages

Documentation

GET /docs

Returns the HTML of OpenAPI (swagger) documentation

Health

GET /health

Returns the health of the API

Context-Free Analysis

Context-free sentiment analysis ignores contextual variations and instead relies primarily on keyword-based or rule-based detection of sentiment. It typically uses predefined lexicons or bag-of-words models to classify text as positive, negative, or neutral, without considering grammar, irony, or surrounding context. Although simpler and faster, it can lead to misinterpretation when word meanings shift with context.

Single Comment Analysis

POST /cf/single
{
  "api-key": "your-api-key",
  "comment": "text to analyze"
}

Batch Comment Analysis

POST /cf/batch
{
  "api-key": "your-api-key",
  "comment-list": ["comment1", "comment2", "..."]
}

Context-Based AnalysisComing Soon

Context-based sentiment analysis takes into account contextual variations and surrounding context to determine sentiment. It often uses machine learning models or rule-based systems to analyze text, taking into account grammar, irony, and surrounding context. This approach is more accurate but can be more complex and slower.

We are raising funds for providing support for context-based analysis because it requires powerful infrastructure. If you want to invest, please mail me at social.sakshamjoshi@gmail.com