Response Structure

This page provides a detailed understanding of the response structure to expect from the streaming api.

The table below explains the structure of responses from the Bodhi API, detailing the meaning of each field and helping you better understand the data returned from the API.

{
  "call_id": "<uuid>",
  "segment_id": <int>,
  "eos": <boolean>,
  "type": "<string>",
  "text": "<string>",
  "segment_meta": {
    "tokens": [],
    "timestamps": [],
    "start_time": <float>,
    "confidence": <float>,
    "words": [
      {
        "word": "<string>",
        "confidence": <float>
      }
    ]
  }
}

Field Descriptions

Key	Description
call_id (string)	Unique identifier associated with every streaming connection
segment_id (string)	Integer associated with every speech segment during the entire active socket connection
eos (bool)	Marks the end of the streaming connection when "eos" is true.
type (string)	Possible values: "partial" | "complete" partial Partial transcript corresponding to every streaming audio chunk complete Complete/final transcript generated for each speech segment Generated once per segment_id i.e., when the speech segment end is reached
text (string)	The transcript that has been processed thus far.
segment_meta (object)	tokens: Array of strings representing individual text pieces (or "tokens") recognized from the audio. Tokens may include words or parts of words. timestamps: Array of numerical values indicating when each token was detected in the segment/sentence (in seconds). Each timestamp aligns with the tokens array, so the i-th timestamp represents the time at which the i-th token was spoken. Useful for measuring latency. start_time: Starting point (in seconds) of the current segment in the overall audio timeline. confidence: Segment level confidence. Float between 0 and 1. words: Array of word level objects (only populated when type is complete). word: The recognised word. confidence: Float value between 0.0 and 1.0 representing the model’s confidence in the recognized word.