Use the following procedure to retrieve task results from IDOL Speech Server.
To retrieve task results
Send the GetResults
action, and set one of the following parameters:
Token
|
The action token for the task. |
File
|
The results file that you want to return. Note: If you set the |
You can set additional parameters. (The exact parameters available depend on the task type. For more information, see the IDOL Speech Server Reference.)
|
The number of results to return, starting at the beginning of the results file. |
Format
|
The results format. Set this parameter to override the default output format for a task. |
Label
|
The label of the output file to return, if the task produces more than one file. By default, IDOL Speech Server returns the results file specified in the task's |
Last
|
The number of results to return, starting from the end of the results file. |
NBest
|
The number of highest-scoring results to return. |
Start
|
The position of the first result to return. |
StartTime
|
The time of the first result to return. You can use this parameter for output from the stt , lid , and lib modules. |
For example:
http://localhost:13000/action=GetResults&Token=MTAuMS4zLjgyOjEzMDAwOkFERFRBU0s6LTEyOTk0OTU3Ng==
http://localhost:13000/action=GetResults&File=C:\softsoundserver\results\out.ctm
IDOL Speech Server tasks produce results in different file formats. You can change the format that results are displayed in whenever you use the GetResults
action to retrieve results. To change the display format, set the Format
parameter in the action. The value that you can set depends on the task and the results file format.
For all tasks and file formats, you can set Format
to one of the following values:
bin
|
Binary data read directly from disk. |
raw
|
XML that contains text lines directly from the results file. |
txt
|
Plain text read directly from disk. |
For any task that generates an XML file, the following value is also available:
xml
|
XML data read directly from disk. |
For some tasks, other values are also available. These tasks must generate CTM files, otherwise the values are unavailable.
Task | Parameter value | Description |
---|---|---|
Audio fingerprint identification ( |
afp
|
XML derived from an AFP CTM file. |
Language identification ( |
lid
|
XML derived from a language identification CTM file. |
Boundary detection-based language identification ( |
lib
|
XML derived from a language identification boundary CTM file. |
Speech-to-text ( |
stt
|
XML derived from a speech-to-text CTM file. |
|