Creating a Voice Profile

A voice profile provides the Speakeasy engine with speech samples in order to get more positive results during speech-to-text. You can create a voice profile from the TE Client after Speakeasy is installed.

Before you create a voice profile, you should have the grammar files configured. When you configure grammar files, you select the specific words and phrases that users will train on to create voice profiles. For more information on selecting the terms for voice profiles, see Specifying Voice Profile Training Options.

Once you have created a voice profile, you can use a script to implement the profile. For an example script that allows you to enter the username and use a voice profile, see Using a Voice Profile.

If you are using an Avalanche Enabler and mobile device server, voice profiles are archived on the mobile device server. If you enter the same username on a different device, the TE Client will automatically retrieve the voice profile from the mobile device server and use it.

To create a voice profile:

1  Ensure that your headset is connected.

2  From the TE Client, click Options > Configure > Microphone.

3  The Options dialog box appears. Click User Training.

4  The Language dialog box appears. Select the language you want to use from the drop-down menu and click OK.

5  The User Name dialog box appears. Type the user name in the text box and click OK.

6  The User Training dialog box appears. Speakeasy vocally prompts you to say the words and phrases selected for user training when the grammar files were configured. You are prompted to say each word or phrase twice. After all the selected terms have been recorded, the profile is saved.

If you only want to train for some of the available terms, stop the training and select the checkboxes next to the terms you want to train for. Then click Start.

If you create a voice profile and later add a grammar file that requires additional training, you must use this process again in order to complete the training.