Microsoft Speech SDK
SAPI 5.1
ISpPhrase::GetText
ISpPhrase::GetText retrieves elements from a text phrase.
HRESULT GetText(
ULONG ulStart,
ULONG ulCount,
BOOL fUseTextReplacements,
WCHAR **ppszCoMemText,
BYTE *pbDisplayAttributes
);
Parameters
- ulStart
- [in] Specifies the first element in the text phrase to retrieve.
- ulCount
- [in] Specifies the number of elements to retrieve from the text phrase.
- fUseTextReplacements
- [in] Boolean value that indicates if replacement text should be used. An example of a text replacement is saying "write new check for twenty dollars" and retrieving the replaced text as "write new check for $20". For more information on replacements, see the SR Engine White Paper.
- ppszCoMemText
- [out] Address of a pointer to the data structure that contains the display text information. It is the caller's responsibility to call ::CoTaskMemFree to free the memory.
- pbDisplayAttributes
- [out] Address of the SPDISPLAYATTRIBUTES enumeration that contains the text display attribute information. Text display attribute information can be used by the application to display the text to the user in a reasonable manner. For example, speaking "hello comma world period" includes a trailing period, so the recognition might include SPAF_TWO_TRAILING_SPACES to inform the application without requiring extra text processing logic for the application.
Return values
Value | Description |
---|---|
S_OK | Function completed successfully. |
S_FALSE | A phrase that does not contain text or ppszCoMemText is NULL. |
E_INVALIDARG | One or more parameters are invalid. |
E_POINTER | Invalid pointer. |
E_OUTOFMEMORY | Exceeded available memory. |
Remarks
The text is the display text of the elements for the phrase and constructs a text string created by CoTaskMemAlloc by applying the pbDisplayAttributes of each SPPHRASEELEMENT.
Example
The following code snippet illustrates the use ISpPhrase::GetText to retrieve parts of the recognized phrase.
HRESULT hr = S_OK; // ... obtain a recognition result object from the recognizer... // get the recognized phrase object hr = cpRecoResult->GetPhrase(&pPhrase); // Check hr // get the phrase's entire text string, including replacements hr = pPhrase->GetText(SP_GETWHOLEPHRASE, SP_GETWHOLEPHRASE, TRUE, &pwszText, NULL); // Check hr // get the phrase's first 2 words, excluding replacements hr = pPhrase->GetText(pPhrase->Rule.ulFirstElement, 2, FALSE, &pwszText, NULL);