Speakeasy Settings

Opens the speaker volume calibration wizard.

Speaker setting for use on Motorola/Symbol mobile devices.

Indicates the sampling frequency.

Displays the full name of the language currently being used.

Displays the three-letter abbreviation of the language currently being used.

Indicates the pitch level of spoken text.

Indicates the speed level.

Indicates how text should be separated.

Indicates the name of the voice that is currently selected.

Indicates the sound level.

This value affects the trade-off between CPU load, memory requirements, and accuracy.

This feature allows the engine to automatically increase and decrease the microphone input volume.

If the confidence value for a result is below this value, then a negative acknowledgement beep will not be played.

Opens the microphone calibration wizard.

Sets how long the user is expected to remain silent during a quick microphone calibration.

Indicates the minimum difference in confidence required between the top two speech-to-text results for the top result to be accepted.

Use this to get the confidence value along with the speech-to-text result.

Indicates the action the speech engine should take before attempting to determine what the user is saying.

Tells the speech engine the distance between the user and the microphone.

Indicates the minimum duration (in ms) of speech before speech detection is activated.

Indicates the speech detection sensitivity.

Indicates the milliseconds of silence used to indicate the user is done speaking.

Indicates the amount of energy the microphone input must have before the speech detection is activated.

Indicates the total milliseconds for the engine to continue collecting results following the last result or timeout.

Displays the full name of the language currently being used.

Displays the three-letter abbreviation of the language currently being used.

Creates a Speech-to-Text log file in the root folder of the device.

Sets the engine to log speech-to-text attempts as .wav files.

If set to 1, the speech-to-text engine will create a log file in the root folder of the device.

Sets the number of terms the engine will examine closely for the best match.

Causes the speech engine to save the current engine state for use later.

Determines how aggressively the microphone input is collected and speech analysis is performed.

Indicates the action the speech engine should take when returning a grammar result.

Modifies engine adaptation speed and/or saved engine information.

Indicates the total milliseconds for the speech engine to wait for a valid response before reverting back to the last saved state.

Causes a sound to play for result recognition.

Increases the threshold for saving a new engine state as time progresses.

Indicates the total milliseconds for the speech engine to wait before saving the next current state.

Directs the speech engine to save the state if the result confidence is greater than the result confidence for stt_threshold and stt_save_threshold combined.

When uploading or downloading user training data, the value for this setting is how long (in seconds) the Client will wait for a response from the Avalanche server.

Displays the size of the speech-to-text engine being used.

Indicates how the speech engine should interpret special sounds.

Indicates the minimum amount of confidence for the most-likely result that will be accepted.

Indicates the total milliseconds (ms) for the system to wait before responding to the speaker.

Sets a buffer to check if the engine is processing speech.

Enables support for Word IDs (the !id directive) in grammar files.