AddSpeakerAudio
Adds one or more audio samples to a speaker.
TIP: You can use this action to add audio samples for training (with training=true
) and for estimating thresholds (with training=false
). You only need to add samples for estimating thresholds if the audio you want to process will contain unknown speakers who are not in the database. You should provide different audio samples for training and for estimating thresholds.
Type: synchronous
Parameter | Description | Required |
---|---|---|
audiodata
|
The audio data to add. Files must be uploaded as multipart/form-data. For more information about sending data to Media Server, refer to the Media Server Administration Guide. | Set this or audiopath |
audiolabels
|
A comma-separated list of labels to identify the audio samples that you are adding (maximum 254 bytes for each label). Every audio sample added to the same speaker must have a unique label, so the number of labels must match the number of samples provided using either audiodata or audiopath . If you do not set this parameter, Media Server generates labels automatically. |
No |
audiopath
|
A comma-separated list of paths to the audio files to add. The paths must be absolute, or relative to the Media Server executable file. | Set this or audiodata |
database
|
The name of the database that contains the speaker. | Yes |
identifier
|
The identifier of the speaker to add the audio samples to. | Yes |
training
|
A Boolean value (default true ) that specifies whether to use the audio sample(s) for training. To add audio samples for estimating thresholds, set this parameter to false . |
No |
Example
The following example adds audio samples to a speaker, for training the speaker model, by specifying the paths of the audio files with the audiopath
parameter:
/action=AddSpeakerAudio&database=news &identifier=JohnSmith &audiopath=sample1.wav,sample2.wav &audiolabels=sample1,sample2 &training=true
The following example adds audio samples, for estimating thresholds, by sending the audio data:
curl http://localhost:14000/action=AddSpeakerAudio -F database=news -F identifier=JohnSmith -F audiodata=@sample3.wav,sample4.wav -F audiolabels=sample3,sample4 -F training=false
Response
<autnresponse> <action>ADDSPEAKERAUDIO</action> <response>SUCCESS</response> <responsedata></responsedata> </autnresponse>
See Also
After adding sufficient audio samples for a speaker, you can run the action BuildSpeaker.
If you expect the audio you want to process to contain unknown speakers (the speakers are an open set) you should also provide audio for some unknown speakers using the action AddUnknownSpeakerAudio.