Recognizer Properties

The following properties may be specified via the configuration file. They may also be overridden via MRCP headers, see Specifying Recognizer Properties via MRCP Headers below.

 

choose_model

 

If this property is set to 1, then the client will decide which acoustic model is most appropriate for the server to use, based on a frequency analysis of the speaker's voice.  Otherwise, two decodes will be done simultaneously, and an answer will be selected based on which model had better "coverage" for the speaker's voice. Default: 0

 

enable_lattice_scoring

 

The lattice based confidence score is a slightly slower, but more accurate confidence score.  Default: 1.

 

end_of_speech_delay

 

Deprecated in favor of speech_complete_timeout, the standard MRCP parameter that works in the same way. See Streaming Media Properties. Default: 500

 

wind_back_time

 

This parameter is the length of audio wound back at the beginning of voice. It helps in the situation of weak speech onset. The resolution of this parameter is 1/8 sec,i.e. 125ms, which means setting this value to 249ms is the same as setting it to 125. It is specified in milliseconds.

Default: 250

 

initial_silence_trimmed

 

This parameter tells the Engine's voice activity detection technology whether the audio stream contains leading silence or not. By default, the Engine expects to receive an audio stream that contains silence before speech. If your hardware is trimming the silence before sending the Engine audio, you should set this to 1.

Default: 0

 

burst_thrsld

 

The FVAD algorithm triggers barge-in only after it has observed the duration of voice longer than this threshold. This threshold helps preventing bursting noise from triggering barge-in. It is specified in milliseconds. The default value is 100.

 

barge_in_timeout

 

If no speech was detected in the time frame specified by this property, then the recognition request is cancelled. It is specified in milliseconds Default: 10000

 

end_of_speech_timeout

 

After barge-in, if it did detect end-of-speech in the time frame specified by this property, then the recognition request is cancelled. It is specified in milliseconds Default: 10000

 

Specifying Recognizer Properties via MRCP Headers

 

Specifying the above parameters in the MRCP headers is optional, but if specified, it must be specified as follows. NOTE: semicolon (";" ) is used as the delimiter.

 

For Example:

 

Vendor-Specific: com.lumenvox.barge-in-lvl=30;com.lumenvox.noise-floor=200\r\n

 

This header-field may be specified in RECOGNIZE or SET-PARAMS method during an MRCP session. Following are the header field names to be used to modify the default values.

 

com.lumenvox.choose-model

com.lumenvox.lattice-score

com.lumenvox.end-of-speech-delay

com.lumenvox.noise-floor

com.lumenvox.wind-back-time

com.lumenvox.initial-silence-trimmed

com.lumenvox.barge-in-timeout

com.lumenvox.end-of-speech-timeout

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


Complete Help Topic List | Speech Engine Product Information