Browse
 
Tools
Rss Categories

Speech Port Properties

Reference Number: AA-01052 Views: 11641 0 Rating/ Voters

Various Speech Port Properties can be set via the API at a Scope of the client level or the port level. In the C API they are set through LV_SRE_SetPropertyEx. In the C++ API, they are set through SetPropertyEx at the port or client level, or SetClientPropertyEx at the client level.SetClientPropertyEx is a static function that allows for client level properties to be applied without opening a port.

The properties that were set can be queried with the following functions:

C API

C++ API

Network Properties


PROP_EX_SRE_SERVERS


Description: This property sets which Speech Engine servers are used for processing decodes. The value is a string containing IP addresses and optional ports separated by semicolons. For instance, "127.0.0.1;10.0.0.1:5721" specifies a server at 127.0.0.1 using the default port of 5730, and a server at 10.0.0.1 using the port 5721.

Scope: Client

Value Type: PROP_EX_VALUE_TYPE_STRING

Possible Values: Strings of IP addresses.

Default Value: "127.0.0.1:5730"


PROP_EX_LICENSE_SERVERS


Description: This property sets which License server is used to check for available licenses. The value is a string containing IP addresses and optional ports separated by semicolons. For instance, "127.0.0.1;10.0.0.1:5721" specifies a server at 127.0.0.1 using the default port of 7569, and a server at 10.0.0.1 using the port 5721.

Scope: Client

Value Type: PROP_EX_VALUE_TYPE_STRING

Possible Values: Strings of IP addresses.

Default Value: "127.0.0.1:7569"

Pre-Processing Properties


PROP_EX_TRIM_SILENCE_VALUE


Description: Sets how aggressively silence is trimmed from the audio during pre-processing. Values range from 0 to 1000; lower values corresponding to more aggressive trimming.

Scope: Port, Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Possible Values: 0 - 1000 

Default Value: 970


PROP_EX_NOISE_REDUCTION_ENABLE


Description: Specifies the noise reduction algorithm used. For most users the default noise reduction algorithm should work best. For certain noise conditions the Alternate noise reduction algorithm has shown better results. Hence, advanced users can try switching the algorithm to see if it improves their performance in noisy conditions. The Adaptive noise reduction algorithm works best only when the noise is constantly changing such as car or highway noise. For more stationary noises like fan noise, the default algorithm will show the best performance

Scope: Port, Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Possible Values:

  • NOISE_REDUCTION_NONE
  • NOISE_REDUCTION_DEFAULT
  • NOISE_REDUCTION_ALTERNATE
  • NOISE_REDUCTION_ADAPTIVE

Default Value: NOISE_REDUCTION_DEFAULT

Decode Properties


PROP_EX_MAX_NBEST_RETURNED


Description: Specifies the maximum number of n-best results to be returned by the Engine.

Scope: Port, Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Possible Values: Number of n-best results. This should be an integer value >= 0

Default Value: 1


PROP_EX_DECODE_TIMEOUT


Description: In a non-blocking decode, this is the timeout value, in milliseconds, used by LV_SRE_WaitForDecode and LVSpeechPort::WaitForDecode functions. In blocking decode, this is the time to wait until the decode times out and returns an error from LV_SRE_Decode and LVSpeechPort::Decode.

Please note that TIMEOUT_INFINITE should not be specified when calling any SetPropertyEx functions - it is a reserved value used for WaitForEngineToIdle calls only.

Scope: Port, Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Possible Values: Time in milliseconds. This should be a positive integer value between 1 and 2147483647 (approximately 600 hours), although these extremes should be avoided

Default Value: 20000


PROP_EX_LOAD_GRAMMAR_TIMEOUT


Description: Specifies how long, in milliseconds, the client should wait for a grammar to load. If the timeout is reached before the grammar is loaded, the LoadGrammar function returns error code -37, LV_LOAD_GRAMMAR_TIMEOUT.

Please note that TIMEOUT_INFINITE should not be specified when calling any SetPropertyEx functions - it is a reserved value used for WaitForEngineToIdle calls only.

Scope: Port, Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Possible Values: Time in milliseconds. This should be a positive integer value between 1 and 2147483647 (approximately 600 hours), although these extremes should be avoided

Default Value: 200000


PROP_EX_ACOUSTIC_MODEL_RESOLUTION


Description: An enumerated type that specifies the resolution of acoustic model that would be used in decode. Higher resolution models may improve accuracy at the cost of speed.

Scope: Port, Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Possible Values:

  • PROP_EX_LOW_RESOLUTION_ACOUSTIC_MODEL
  • PROP_EX_MIDDLE_RESOLUTION_ACOUSTIC_MODEL
  • PROP_EX_HIGH_RESOLUTION_ACOUSTIC_MODEL

Default Value: PROP_EX_LOW_RESOLUTION_ACOUSTIC_MODEL


PROP_EX_SPEED_VS_ACCURACY


Description: An integer between 0-100 specifying how much accuracy will be valued in an SRE decode, at the cost of speed. Lower values correspond to faster decodes; 100 means maximum accuracy.

Scope: Port, Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Possible Values: 0-100

Default Value: 50

Grammar Properties


PROP_EX_STRICT_SISR_COMPLIANCE


Description: Controls whether LumenVox will strictly implement the final SISR 1.0 standard for adding tags to grammars. This value must be changed to 0 in order to run in a backward-compatible mode able to read the older tag format.

Scope: Port, Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Possible Values:

  • 0 (disabled)
  • 1 (enabled)

Default Value: 1


PROP_EX_BUILTIN_GRAMMAR_LANGUAGE


Description: Accepts a string that specifies the language that builtin grammars default to. See Built-in Grammars for information on working with built-in grammar languages.

Scope: Port, Client

Value Type: PROP_EX_VALUE_TYPE_STRING

Possible Values:

  • "en-US"
  • "es-MX"

Default Value: "en-US"

Licensing Properties


PROP_EX_LICENSE_TYPE


Description: The license type sets the default license type to used when opening any port on that client. The value options are PROP_EX_LICENSE_VOXLITE, PROP_EX_LICENSE_SPEECHPORT. If the value is set to PROP_EX_LICENSE_VOXLITE, the client will get the license from the Lite license pool (these licenses only allow up to 500 vocabulary items per recognition). If the value is set to PROP_EX_LICENSE_SPEECHPORT, the client will get the license from the Full license pool. By default, the client will auto-pick the license; it will use up Full licenses before using Lite licenses. Obviously only licenses that are available may be obtained

Scope: Client

Value Type: PROP_EX_VALUE_TYPE_STRING

Possible Values:

  • PROP_EX_LICENSE_VOXLITE
  • PROP_EX_LICENSE_SPEECHPORT
  • PROP_EX_LICENSE_AMD
  • PROP_EX_LICENSE_SLM

Default Value: "AUTO"

Logging Properties


PROP_EX_SAVE_SOUND_FILES


Description: Controls whether the application will save off .callsre files used with the LumenVox Speech Tuner. Turn this on to capture audio and more information related to each decode. 

  • SAVE_SOUND_FILES_NONE - Deactivates saving of .callsre files. 
  • SAVE_SOUND_FILES_BASIC - Saves basic information in .callsre files. This includes the audio from when BARGE_IN occurred up to the END_OF_SPEECH. 
  • SAVE_SOUND_FILES_ADVANCED - In addition to the information stored with the SAVE_SOUND_FILES_BASIC setting, when speech is streamed in to the Voice Activity Detection module, all of the collected data prior to a StreamCancel command is saved. This is usually when a NO_INPUT or TIMEOUT event occurs. This option collects data only when there is something to debug, which offers a balance between disk usage and retaining important information for debugging and is particularly useful when diagnosing NO-INPUT problems. 
  • SAVE_SOUND_FILES_ALL - Used to collect all streamed data in all cases. This option collects all streamed data, whether there was a NO_INPUT or TIMEOUT event or not, including untrimmed audio from 'good' decodes. This option can be useful in diagnosing barge-in problems as well as other potential issues

Scope: Port, Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Possible Values:

  • SAVE_SOUND_FILES_NONE
  • SAVE_SOUND_FILES_BASIC
  • SAVE_SOUND_FILES_ADVANCED
  • SAVE_SOUND_FILES_ALL

Default Value: SAVE_SOUND_FILES_NONE


PROP_EX_LOGGING_VERBOSITY


Description: Controls the verbosity of event logging. This can be used to increase or decrease the amount of log events that are generated. Note that increasing logging verbosity uses more CPU, and should therefore be avoided wherever possible in production systems where optimal performance is critical. 

Scope: Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Possible Values:

  • 1 - Minimal logging. Logs only errors and critical issues.
  • 2 - Medium logging. Logs all non-debug information as events occur.
  • 3 - Maximum logging. Logs all types of events. This will include any and all informational and debugging activity.

Default Value: 1


PROP_EX_SECURE_CONTEXT


Description: Controls the suppression of event logging. This can be used to suppress potentially sensitive data from logged events that are generated. This setting may be used with either ASR or TTS clients. When enabled, this setting will prevent logging of results or any other data to either the log files, or the callsre files. Wherever sensitive data would have appeared, this will be replaced with the word _SUPPRESSED, so indicate that data was suppressed. In the case of recorded audio, these will not be placed into the callsre file when this secure_context mode is enabled.

This functionality was introduced with LumenVox version 11.0.300 (November 2012)  as part of our ongoing enhancements to support secure application development.

Scope: Port, Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Possible Values:

  • 0 - Disabled. Regular logging will occur as normal
  • 1 - Secure context mode enabled. Logging of potentially sensitive data will be suppressed

Default Value: 0


PROP_EX_CONFIDENCE_THRESHOLD


Description: Records the current confidence threshold setting value of the associated speech application. Assigning this value has not functional effect within the speech port, but is recorded along with other decode related settings when creating response (.callsre) files.  Having this value available can be extremely helpful when tuning applications, since the LumenVox Speech Tuner has the ability to calculate the optimal threshold value and compare it against this record value if it is available. We recommend setting this value whenever possible.  The LumenVox Media Server will automatically assign Confidence-Threshold values as defined by the active RECOGNIZE request.

This functionality was introduced with LumenVox version 12.1.100 (August 2014)

Scope: Port, Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Possible Values:

  • 0 - 1000

Default Value: 0



Deprecated or Non-Functional Properties


PROP_EX_LIC_SERVER_HOSTNAME


Description: A string specifying the IP address of the machine hosting the LumenVox license server. This will be removed.

Scope: Client

Value Type: PROP_EX_VALUE_TYPE_STRING

Status: Deprecated since 10.0

Possible Values: Strings of IP addresses.

Default Value: "127.0.0.1"

PROP_EX_LIC_SERVER_PORTNUM


Description: An integer specifying the port number used by the LumenVox license server. This will be removed.

Scope: Client

Value Type: PROP_EX_VALUE_TYPE_INT, PROP_EX_VALUE_TYPE_INT_PTR

Status: Deprecated since 10.0

Possible Values: Port Number

Default Value: 7569

PROP_EX_DECODE_THREAD_PRIORITY


Description: Specifies the priority for the decode thread on the server. Higher priority threads will be processed before other applications on the same machine. Adjusting this parameter is useful for troubleshooting load issues on a machine.

Status: Deprecated and non functional since 9.5

PROP_EX_DECODE_OPTIMIZATION


Description: Sets whether the Engine should prioritize speed or accuracy while performing a decode. By default, it will switch between modes depending on how much load is happening on the server machine; when the server becomes too busy it will begin optimizing speed over accuracy.

Status: Deprecated and non functional since 9.5

PROP_EX_LANGUAGE


Description: A string specifying the default language to be used for the deprecated Concept/Phrase type grammars. This is no longer supported.

Status: Deprecated and non functional since 10.0

PROP_EX_NOISE_REDUCTION_AUTO_ADAPTIVE


Status: Deprecated and non functional since at least 9.5

PROP_EX_NOISE_REDUCTION_ALPHA


Status: Deprecated and non functional since at least 9.5

PROP_EX_CHOOSE_MODEL


Status: Deprecated and non functional since at least 9.5

PROP_EX_SET_SERVER_IP


Status: Deprecated and non functional since at least 9.5

PROP_EX_SET_SERVER_PORT


Status: Deprecated and non functional since at least 9.5

PROP_EX_SEARCH_BEAM_WIDTH


Status: Deprecated and non functional since at least 9.5

PROP_EX_CONCEPT_REPETITION_MIN


Status: Deprecated and non functional since at least 9.5

PROP_EX_CONCEPT_REPETITION_MAX


Status: Deprecated and non functional since at least 9.5

PROP_EX_ENABLE_LATTICE_CONFIDENCE_SCORE


Status: Deprecated and non functional since at least 9.5

PROP_EX_MOD_SEL_LOW_THLD


Status: Deprecated and non functional since at least 9.5

PROP_EX_MOD_SEL_HIGH_THLD


Status: Deprecated and non functional since at least 9.5

PROP_EX_FLAT_SEARCH


Status: Deprecated and non functional since at least 9.5

PROP_EX_TREE_SEARCH


Status: Deprecated and non functional since at least 9.5

PROP_EX_WORDSIZE_AUTOPICK_SEARCH


Status: Deprecated and non functional since at least 9.5

See Also