Tracks recognition requests; this a client side table, in the sense that this is the kind of information you would send to a recognition platform to get a rec result. Typically, the request will include the grammar, perhaps a language or acoustic model request, and administrivia like session and utterance identifiers, search parameters, etc.
| Column Name | Data Type | Data Length | Default Value | See Also |
|
RequestID |
INTEGER |
N/A |
N/A |
N/A |
|
InteractionID |
INTEGER |
N/A |
N/A |
|
|
EachDigit |
INTEGER |
N/A |
N/A |
N/A |
|
DecodeFlags |
INTEGER |
N/A |
N/A |
N/A |
|
stLanguageModel |
INTEGER |
N/A |
N/A |
|
|
OOV |
INTEGER |
N/A |
N/A |
N/A |
|
StandardGrammar |
INTEGER |
N/A |
N/A |
N/A |
|
PitchResult |
INTEGER |
N/A |
N/A |
N/A |
|
NoiseReduction |
INTEGER |
N/A |
N/A |
N/A |
RequestID
| Foreign key | Reference table | Reference key |
|
InteractionID |
LVST_CallInteractions |
InteractionID |
|
stLanguageModel |
LVST_StringTable |
StringID |
RequestID
Description:Primary key for the table
Constraint Name:N/A
Constraint Description:N/A
InteractionID
Description:Foreign key into tblLVST_CallInteractions. Identifies the interaction this request belongs to.
Constraint Name:N/A
Constraint Description:N/A
EachDigit
Description:Controls whether the returns on a digit decode (using standard grammar digits) returns a collapsed version (i.e. all digits returned as a single string), or returns each digit separately (i.e. each digit returned as a separate string). For example, an utterance of 'one two three four five' can be either '12345' (EachDigit = 0), or '1', '2', '3', '4', '5' (EachDigit = 1). This field is LumenVox-specific.
Constraint Name:N/A
Constraint Description:N/A
DecodeFlags
Description:Controls certain decode options (such as USE_SRGS). The decode flags are constructed using bitwise ||, and checked using bitwise AND. This field is LumenVox-specific.
Constraint Name:N/A
Constraint Description:N/A
stLanguageModel
Description:Foreign key to tblLVST_StringTable. The language to use for the decode. For LumenVox, this is the name of the language model file on disk, without ".model". Despite the confusing name, this is *not* a statistical language model, this is a grouping, by language (for example, 'AmericanEnglish'), of a set of acoustic models.
Constraint Name:N/A
Constraint Description:N/A
OOV
Description:Controls whether out-of-vocabulary filtering is turned on or not. This field is LumenVox-specific.
Constraint Name:N/A
Constraint Description:N/A
StandardGrammar
Description:Controls the kind of standard grammar (for LV_Concepts *only*; with SRGS, use the built-in grammars) to use for the decode. This field is LumenVox-specific.
Constraint Name:N/A
Constraint Description:N/A
PitchResult
Description:Indicates the acoustic model in the language to use for recognition, if pitch detection is requested.
Constraint Name:N/A
Constraint Description:N/A
NoiseReduction
Description:Controls whether noise reduction is enabled or not. This field is LumenVox-specific.
Constraint Name:N/A
Constraint Description:N/A
Constraint Name:N/A
Constraint Description:N/A
Columns related:N/A