Full Transcript

The Full Transcript feature provides a cumulative transcript that accumulates all transcription text received up to the current point in the session. This is useful for maintaining a complete transcript of the entire conversation or audio stream.

Enabling Full Transcript

Add full_transcript=true to your WebSocket connection query parameters when connecting to the Pulse STT WebSocket API. The default is false.

Real-Time WebSocket API

1
const url = new URL("wss://waves-api.smallest.ai/api/v1/pulse/get_text");
2
url.searchParams.append("language", "en");
3
url.searchParams.append("encoding", "linear16");
4
url.searchParams.append("sample_rate", "16000");
5
url.searchParams.append("full_transcript", "true");
6
7
const ws = new WebSocket(url.toString(), {
8
  headers: {
9
    Authorization: `Bearer ${API_KEY}`,
10
  },
11
});

Output Format & Field of Interest

The full_transcript field contains the complete transcription text accumulated from the beginning of the session. This field is only included in responses where full_transcript=true query parameter is set AND is_final=true, ensuring you receive the complete transcript only when a segment is finalized.

Sample Response

1
{
2
  "session_id": "sess_12345abcde",
3
  "transcript": "How are you doing today?",
4
  "is_final": true,
5
  "is_last": false,
6
  "full_transcript": "Hello, my name is John. How are you doing today?",
7
  "language": "en",
8
  "languages": ["en"]
9
}

Response Fields