The text normalizer converts numbers, dates, punctuation, and so on, in a transcript file to a standard form that is recognized by the speech-to-text process. You must normalize text before you run the transcript checking and alignment tasks, which use the align
module.
Mode | Input | Output | Description |
---|---|---|---|
Does not accept any input streams or produce any output streams. Inputs and outputs are set as parameters. |
Example:
output <- textnorm (_, input)
DropBracketedText | PgfFile | TXTFileOutLabel |
DropComments | TempDir | WordPerLine |
KeepMetadata | TXTFileIn* | |
Lang* | TXTFileOut* |
*You must set the value of these configuration parameters by setting the corresponding action parameters.
|