Audio Matching Results

The following XML shows a single record produced by audio matching.

<record>
    <timestamp>
       ...
    </timestamp>
    <trackname>AudioMatchTask.Result</trackname>
    <AudioMatchResult>
        <id>4d69390f-a8c4-4c5d-a0b0-705a3f98aa9b</id>
        <recognitionOffset>0</recognitionOffset>
        <identity>
            <identifier>example</identifier>
            <database>music</database>
            <confidence>60</confidence>
            <metadata>
                <item>
                    <key>key1</key>
                    <value>value1</value>
                </item>
                <item>
                    <key>key2</key>
                    <value>value2</value>
                </item>
            </metadata>
        </identity>
    </AudioMatchResult>
</record>

The record contains the following information:

  • id - a unique identifier for the match.
  • recognitionOffset - a non-zero value indicates that only part of the clip was found in the ingested audio. For example, if you add a 60-second audio clip to your training database, and the ingested audio contains the last 25 seconds of the clip but not the first 35 seconds, the recognitionOffset element will contain the value 35000 (the value is given in milliseconds).
  • identity - describes the clip that was recognized. This element contains the following information:

    • identifier - the identifier of the clip that was recognized.
    • database - the name of the database that contains the audio clip.
    • confidence - the confidence score.
    • metadata - any custom metadata associated with the clip. (You can add custom metadata to audio clips in your training database).