Skip to main content
Deepgram's Docs

Search documentation

Type to search this documentation.

On this pageOverview

Topic Detection

Deepgram API Playground

Try this feature out in our API Playground.

topics boolean Default: false

English (all available regions)

Topic Detection accepts an input text, divides it into a list of segments comprised of sections of the text, and identifies key topics found within each segment.

JSON
 "results": {
    "topics": {
      "segments": [
        {
          "text": "Hi I'm calling to get a refund on my recent purchase. Sure I'd be happy to help you with that. What was the number for you order?",
          "start_word": 0,
          "end_word": 26,
          "topics": [
            { "topic": "Refund", "confidence_score": 0.91318 },
            { "topic": "Order Number", "confidence_score": 0.95342 }
          ]
        },
        {
          "text": "Ok thanks for that. It looks like you made this purchase online, is that correct? Yes I ordered this online on your website a few days ago.",
          "start_word": 45,
          "end_word": 72,
          "topics": [
            {
              "topic": "Online Transacation", "confidence_score": 0.741929
            }
          ]
        }
      ]
    }
  }

The list of topics that can be identified are not a fixed list; this TSLM powered feature is able to generate topics based on the context of the language content in the text. You may also choose to use the optional custom-topic parameter to provide a custom topic you want detected if present within the provided text.

To enable Topic Detection, use the following parameter in the query string when you call Deepgram’s /read endpoint:

topics=true

To analyze text from a file on your computer, run the following curl command in a terminal or your favorite API client.

Bash

To analyze text from a hosted file, run the following curl command in a terminal or your favorite API client. (Try testing it out with the hosted file https://static.deepgram.com/examples/aura.txt)

To tell the model to only return topics from your own custom list of topics, add custom_topic_mode=strict and custom_topic= followed by the list of topics. (Use the URL encoding%20 to represent a space between each word in the list.)

If you want to return your own custom list of topics in addition to Deegpram’s list of topics, set custom_topic_mode=extended and add your custom list.

Bash
Parameter Value Type Description
topics true boolean Enables Topic Detection
language en string The language of your input text (Only English is supported at this time.)
custom_topic ex: animals string Optional. A custom topic you want the model to detect within your input text if present. Submit up to 100.
custom_topic_mode extended, strict string Optional. Sets how the model will interpret strings submitted to the custom_topic param. When strict, the model will only return topics submitted using the custom_topic param. When extended, the model will return its own detected topics in addition to those submitted using the custom_topic param.

When the file is finished processing, you’ll receive a JSON response that has the following basic structure:

JSON
{
  "metadata": {
    "request_id": "c313ae16-2c3b-4c51-87a6-920a8aa1d899",
    "created": "2023-11-28T01:44:27.083Z",
    "language": "en",
    "topics_info": {
      "model_uuid": "ba5b22e4-b39a-4550-a4bc-d8655f5092bc",
      "input_tokens": 22,
      "output_tokens": 4
    }
  },
  "results": {
    "topics": {
      "segments": [
        {
          "text": "Hi I'm calling to get a refund on my recent purchase. Sure I'd be happy to help you with that. What was the number for you order?",
          "start_word": 0,
          "end_word": 26,
          "topics": [
            { "topic": "Refund", "confidence_score": 0.91318 },
            { "topic": "Order Number", "confidence_score": 0.95342 }
          ]
        },
        {
          "text": "Ok thanks for that. It looks like you made this purchase online, is that correct? Yes I ordered this online on your website a few days ago.",
          "start_word": 45,
          "end_word": 72,
          "topics": [{ "topic": "Online Transacation", "confidence_score": 0.741929 }]
        }
      ]
    }
  }
}

The response object values for topics are:

  • segments: The list of segments of text identified by the model as containing notable topics.
  • topic: The name of the topic detected by the model.
  • confidence_score: a floating point from 0 to 1 representing the models confidence in this prediction.

Status 400

If you request Topic Detection with an unsupported language by specifying a language code such as topics=true&language=es or topics=true&detect_language=true where the detected language is unsupported, you will get the error message below.

JSON
{
  "err_code":"INVALID_QUERY_PARAMETER",
  "err_msg":"Request specified unsupported language: <language_name>. Only English is supported.",
  "request_id":"XXXX"
}

Status 400

If the request’s input length exceeded the 150k token rate limit per request, you will get the error message below.

JSON
{
  "err_code": "TOKEN_LIMIT_EXCEEDED",
  "err_msg": "Text input for <api_name> currently supports up to 150K tokens. Please revise your text input to fit within the defined token limit. For more information, please visit our API documentation.",
  "request_id": "XXXX"
}

Status 400

If the request sent contained only the feature parameter (topics) but not the language parameter, you will receive this error.

JSON
{
  "err_code":"INVALID_QUERY_PARAMETER",
  "err_msg":"Failed to deserialize query parameters: missing field `language`",
  "request_id":"XXX"
}
Suggest an edit

Propose a replacement for this page. The site team reviews it before applying any changes.

Export
Documentation menu