SpeechRecognizer

MLModule
genre Other
author Felix Ritter
package FMEwork/Release
dll MLSpeechRecognizer
definition MLSpeechRecognizer.def
keywords speech, NLU, voice

Purpose

Module to perform speech recognition. (macOS-only currently)

Windows

Default Panel

../../../Projects/MLSpeechRecognizer/Modules/mhelp/Images/Screenshots/SpeechRecognizer._default.png

Parameter Fields

Visible Fields

Locale

name: locale, type: String, default: de-DE

Locale representing the language and country you want to use for speech recognition.

Contextual Strings

name: contextualStrings, type: String

A newline seperated list of phrases that should be recognized, even if they are not in the system vocabulary.

Operate without network access

name: requiresOnDeviceRecognition, type: Bool, default: FALSE

A Boolean value that determines whether a request must keep its audio data on the device.

Start

name: start, type: Trigger

Start speech recognition process.

Stop

name: stop, type: Trigger

Stop speech recognition process.

Transcription Intermediate

name: transcriptionIntermediate, type: String, persistent: no

Intermediate results returned for each utterance.

Transcription Delta

name: transcriptionDelta, type: String, persistent: no

Difference (delta) of the current to the last recognized utterance.

Transcription Finished

name: transcriptionFinished, type: String, persistent: no

The complete utterance after a period of silence; speech processing is complete.

Hidden Fields

supportedLocalesList

name: supportedLocalesList, type: String, persistent: no

Comma separated list of language-country codes for which speech recognition is supported.

supportsOnDeviceRecognition

name: supportsOnDeviceRecognition, type: Bool, persistent: no

A Boolean value that indicates whether the speech recognizer can operate without network access.

contextualStringsTextFilePath

name: contextualStringsTextFilePath, type: String

Path to text file which contains a newline seperated list of phrases that should be recognized, even if they are not in the system vocabulary.

endOfUtteranceDelay

name: endOfUtteranceDelay, type: Double, default: 1

Period of silence in seconds required to end an utterance.

isRunning

name: isRunning, type: Bool, persistent: no

A Boolean value that indicates whether the speech recognition process is running.

utteranceReinterpreted

name: utteranceReinterpreted, type: Trigger

Field gets triggered when an utterance gets reinterpreted. Transcription Intermediate is set before and Transcription Delta afterwards.

hadError

name: hadError, type: Trigger

Field gets triggered after an error occurred.

errorDescription

name: errorDescription, type: String, persistent: no

The last error message.