TrainCustomSpeechWord
Adds a word to a custom word database for speech-to-text.
If the word already exists in the custom word database, this action overwrites the word. You can therefore use this action to change the associated base word and weight.
Type: synchronous
Parameter | Description | Required |
---|---|---|
database
|
The name of the database to add the word to. | Yes |
word
|
The word to add. | Yes |
baseword
|
A word that exists in the base language pack, that could appear in the same context or a similar context as the word you are adding. For example, if you are adding a company name, you could use a company name that already exists in the base language pack. Specifying a base word is optional, but provides Media Server with some information about the context in which the custom word is expected to appear. |
No |
weight
|
A multiplier (from 0 to 1, default 0.5) that specifies how likely the word is to appear, relative to the baseword . For example, if the word you are adding is only slightly less likely to appear than the baseword , you might set this parameter to 0.8 . If the word is much less likely to appear you might set this parameter to 0.1 . |
No |
Example
The following example adds the word "AcmeSoft" to the database "words" and instructs Media Server that the word is expected to appear in the same context or a similar context as the word "Microsoft".
/action=TrainCustomSpeechWord&database=words &word=AcmeSoft &baseword=Microsoft &weight=0.7
Response
The following XML is an example response:
<autnresponse> <action>TRAINCUSTOMSPEECHWORD</action> <response>SUCCESS</response> <responsedata></responsedata> </autnresponse>