Batch speaker identification

Learn how to use the Speechmatics API to identify speakers in Batch

For an overview of the feature, see the speaker identification page.

Enrollment

To generate identifiers for a desired speaker, run a speaker diarization enabled transcription on an audio sample where the speaker is ideally speaking alone. You can request the identifiers back from the engine by setting the get_speakers flag in the transcription config:

{
  "type": "transcription",
  "transcription_config": {
    "model": "enhanced",
    "language": "en",
    "diarization": "speaker",
    "speaker_diarization_config": {
      "get_speakers": true
    }
  }
}

When the transcription is done, the speakers identifiers will be attached to the returned transcript:

{
  "results": [
    {
      "alternatives": [
        {
          "confidence": 0.93,
          "content": "Hello",
          "language": "en",
          "speaker": "S1"
        }
      ],
      ...
    },
    {
      "alternatives": [
        {
          "confidence": 1.0,
          "content": "Hi",
          "language": "en",
          "speaker": "S2"
        }
      ],
      ...
    }],
  "speakers": [
    {
      "label": "S1",
      "speaker_identifiers": ["<id1>"]
    },
    {
      "label": "S2",
      "speaker_identifiers": ["<id2>"]
    }]
}

Identification

Once you have generated speaker identifiers, you can provide them in your next transcription job to identify and tag known speakers. This is done through the speakers option in the speaker diarization configuration. All other speaker diarization options remain supported. The speaker_sensitivity setting in speaker_diarization_config adjusts how strongly the system prefers enrolled speakers over detecting new generic ones. Lower values make it more likely to match existing enrolled speakers.

An example configuration is shown below:

{
  "type": "transcription",
  "transcription_config": {
    "model": "enhanced",
    "language": "en",
    "diarization": "speaker",
    "speaker_diarization_config": {
      "speakers": [
        {"label": "Alice", "speaker_identifiers": ["<alice_id1>", "<alice_id2>"]},
        {"label": "Bob", "speaker_identifiers": ["<bob_id1>"]}
      ],
      "speaker_sensitivity": 0.4
    }
  }
}

With the config above, transcript segments should be tagged with "Alice" and "Bob" whenever these speakers are detected, whereas any other speakers should be tagged with the internal labels:

{
  "results": [
    {
      "alternatives": [
        {
          "confidence": 0.93,
          "content": "Morning",
          "language": "en",
          "speaker": "Alice"
        }
      ],
      ...
    },
    {
      "alternatives": [
        {
          "confidence": 0.93,
          "content": "Hi",
          "language": "en",
          "speaker": "S1"
        }
      ],
      ...
    },
    {
      "alternatives": [
        {
          "confidence": 1.0,
          "content": "Morning",
          "language": "en",
          "speaker": "Bob"
        }
      ],
    }]
}