Microsoft Speech SDK SAPI 5.1

---

SPPARSEINFO

SPPARSEINFO is filled in by the speech recognition (SR) engine and sent to SAPI. SAPI uses the information to build a phrase object that can be sent in a hypothesis or recognition (see ISpSREngineSite::ParseFromTransitions).

typedef struct SPPARSEINFO
{
   ULONG          cbSize;
   SPRULEHANDLE   hRule;
   ULONGLONG      ullAudioStreamPosition;
   ULONG          ulAudioSize;
   ULONG          cTransitions;
   SPPATHENTRY   *pPath;
   BOOL           fHypothesis;
   GUID           SREngineID;
   ULONG          ulSREnginePrivateDataSize;
   const BYTE    *pSREnginePrivateData;
} SPPARSEINFO;

Members

cbSize

The size of the SPPARSEINFO structure, as set by the SR engine.

hRule

The handle of the top-level rule for the recognition or hypothesis.

ullAudioStreamPosition

The position in the stream where the recognition started. If downsampling an audio stream, ullAudioStreamPosition will be the byte position within the original stream.

ulAudioSize

The size in bytes of the portion of the stream that is recognized.

cTransitions

The number of word transitions in the pPath array.

pPath

The transition path through the CFG that is recognized.

fHypothesis

TRUE if a hypothesis, false if a final recognition.

SREngineID

The unique identifier of the SR engine. Can be NULL.

ulSREnginePrivateDataSize

The size of private SR engine data that will be sent to the SR engine alternates analyzer. Can be zero.

pSREnginePrivateData

A pointer the private SR engine data (memory) that will be sent to the SR engine alternates analyzer. Can be NULL. SAPI will handle the serialization and marshaling of the data for the SR engine.