Speech

Prev Next

Choose from a range of voice providers. Each provider offers distinct voices to suit your preferences.

Standard Voices

Device voices: Voices that come with the device you own. Some device voices are pre-downloaded and ready to use. Although free to use, certain device voices may require downloading or installing. If you choose to use a device voice, you will first need to select the voice accent.

Online voices: These voices require an internet connection and will use your quota for online features. To check the status of your online features quota, go to Online Features settings.

Once you have selected the voice accent, you can select a voice to preview and select it. If a voice doesn't speak, you can enable it by downloading it in your device settings.

To download a voice on an iOS device:

Go to iOS settings > Accessibility > Spoken content > Voices and download any available voices there.

To download a voice on an Android device:

To find your Text-to-Speech settings, go to the Android settings > System menu > Language and Input > Text-to-Speech options (the location of this setting may vary depending on the device). If Google Text-to-Speech is not pre-installed on your Android device, you can download it from the Google Play store.

Note: Samsung devices come with Samsung Text-to-Speech installed by default in the Text-to-Speech options.


Voices Banking Providers

Explore a variety of voice banking providers to personalize your communication experience. Each provider offers unique features and capabilities.

Acapela

Acapela is an external company providing personalized digital voices from their My-Own-Voice (MOV) voice banking service. Using an Acapela MOV allows users to communicate with a unique personal synthetic voice that is representative of their own voice.

To download an Acapela MOV demonstration voice, open the speech settings of the app, select Acapela as the voice engine and enter credentials and click on the download button.

To create your own Acapela MOV voice, you will need to complete your voice banking with Acapela. To do so you will need to follow these steps:

  • Apply to the My-Own-Voice online service: mov.acapela-group.com/start-now/

  • Record the sentences at your own pace

  • Listen to your voice in the Type and Talk interface

  • When ready, you can decide whether to buy their services.

To download the Acapela MOV voice, go to the Predictable Speech settings, enter your Acapela username and password, and then select download.

ElevenLabs

ElevenLabs allows you to create and use your own custom voice, or a variety of pre-made voices. If you haven’t already created a custom voice and would like one, set up an account with ElevenLabs at elevenlabs.io/. You can record or upload recordings to the ElevenLabs website to create your own custom voice. If you have permanent speech loss caused by a diagnosed disease, ElevenLabs Professional Voice Clones are available for free.

Using your ElevenLabs Voice in Predictable

  1. Go to the ElevenLabs website to generate your unique API key. This is what you will use to connect your voice to Predictable.

  2. Select My Workspace, select API keys, and then select Create a new API key (you can name the API key anything you want).

  3. You will be shown your API key which you should copy and save this right away. You will not be able to view it again.

    Note: If you lose this API key, you can return to this screen and generate a new API key.

  4. Then go to Predictable settings and select Speech. Under “Voice Engines” select ElevenLabs.

  5. Enter or copy/paste the API key that you generated into the input box and then select Connect my voices.

  6. You should now be able to view and select all of your voices from a list.

Using Predictable Offline when you have an ElevenLabs Voice

ElevenLabs are online voices. You will not be able to speak new text when offline. However, here are the options to help in this scenario:

  • Recordings from ElevenLabs are saved in your Predictable history, so you will be able to speak anything you have spoken before with an ElevenLabs voice, even when offline.

  • There is an option in Settings to Download your Grid. You can download your whole grid which will pre-load the ElevenLabs recordings and allow you to continue accessing your Grid, even when offline.

    Note: The Download your Grid option uses a significant amount of your ElevenLabs quota.

ModelTalker

ModelTalker is a speech synthesis software package developed by the Nemours Speech Research Laboratory with the option to use a banked voice. Banked voices are synthetically re-created voices that can be used with Predictable.

To complete your voice banking with ModelTalker you will need to complete and submit this online form: modeltalker.org/registration/registration to create an account with ModelTalker.org and gain access to their voice recording software. You will find more information about how to record your voice on their website: modeltalker.org/getting-set-up-for-recording/.

ModelTalker support contact information: staff@modeltalker.org

To download your custom ModelTalker voice, open the Predictable Speech settings, select ModelTalker as the voice engine, enter your ModelTalker credentials, and then select download.

SpeakUnique

SpeakUnique offers three voice banking services: voice build, voice repair, and voice design. The voice you bank can then be used in Predictable.

If you have questions about the SpeakUnique voice banking services, contact SpeakUnique via their website: speakunique.co.uk/overview/

To download your SpeakUnique voice, open the Predictable Speech settings, select SpeakUnique as the voice engine, enter your SpeakUnique username and password, and then select Download.

The Voice Keeper

The Voice Keeper service is run by Almagu, a synthetic speech company that allows a user to create a synthetic voice for their AAC device by recording their own natural voice.

For more information about how to create your voice with The Voice Keeper, go to thevoicekeeper.com/.

To download your Voice Keeper voice, open the Speech settings in Predictable, select The Voice Keeper as the voice engine, enter your Voice Keeper username and password, and then select Download.


Speaking Rate and Pitch

Use the slider to change the speaking rate and speaking pitch for the voice you choose. Select the Play button to hear a preview of the voice with your settings.

Note: This feature may not work for all voice banking providers.


Speak Button Customizations

You can customize the Speak Button to your preferences. There are three options to choose from for Speak Button functionality: Communication Type, Speak Interruptions, and Speak Interactions.

When the Message Window Speak Button is enabled, selecting the Message Window will activate the Speak Button the way you have it customized. Note that enabling this feature will disable all other possible interactions with the Message Window, such as moving the cursor, copying, pasting, etc.

Communication Type

Choose between "Show" and "Speak" for your Communication Type.

Show

When selected, the text in your message window is displayed backward.

Speak

When selected, the Speak Button will read your message.

Speak Interruption

Choose one of the options to customize how you start and stop your speech.

Stop

Activate the Stop function to stop the entire message being spoken. If you want to speak the same message again, you'll need to start from the beginning rather than resuming from the point where you selected the Stop button.

Pause

Activate the Pause function to pause the message being spoken. Unlike Stop, Pause allows you to continue from the paused point in your message when you select the play button. This provides greater flexibility when managing spoken messages.

Non-stop speak

This will remove Stop or Pause functionality from your Speak Button. Selecting the Speak Button will speak out all text and then it will be unavailable to select again. This can help prevent accidentally selecting the Speak Button a second time and duplicating the spoken message.

Speech Behavior

Toggle on one of the following options to set the way your speech functions.

Clear message window after speaking

Turn on this option to automatically clear the message window after the speech is completed, providing a clean slate for your next communication.

Speak after space

Turn on this option to speak the text in the message window automatically when the spacebar is pressed.

Speak after sentence

Turn on this option to speak as soon as sentence has been completed.

Speak Interaction

Refine how your messages are spoken by customizing the Speak Interactin options. There are several customization options for the Speak Button and the Message Window.

Speak all

The entire message in the Message Window is spoken when the Speak Button is selected.

Speak last sentence

Only the last sentence in the message will be spoken.

Speak unspoken text

Only words that haven't been spoken previously will be spoken.

Speak by paragraph

Speak the first paragraph before stopping. Select Speak by Paragraph again to speak the next paragraph.

Speak by sentence

Speak the first sentence before stopping. Select Speak by Sentence again to speak the next sentence.

Speak after cursor

Speak all text after the current position of your cursor in the Message Window.

Speak before cursor

Speak all text before the current position of your cursor in the Message Window.

Speak current sentence

Speak the sentence your cursor is currently placed in.


Grid Speech Options

Send Phrases to Message Window

Turn on this option to send phrases directly to the message window when you select them.

Speak Phrases on Tap

Turn on this option to speak a phrase aloud when it is selected in the phrase bank.

Speak Categories on Tap

Turn on this option to speak the category name aloud when the category is selected.


Custom Pronunciation

If the synthetic voice is not speaking a word as you would like, you can change the pronunciation of the word using the Custom Pronunciation feature. For example, if the word "cough" is not pronounced as you prefer, use custom pronunciation to specify the desired phonetic spelling (spell it as it sounds), such as "coff."

Customizing the pronunciation of words enhances the clarity and accuracy of the synthetic voice. Explore the options to add, edit, or delete custom pronunciations.

Add a Pronunciation

  1. Enter the display text and provide your desired phonetic pronunciation.  

Edit an Existing Pronunciation

  1. Select the pronunciation you want to change.

  2. Select Edit in the window that appears.

  3. Change the word or phonetic pronunciation as needed.

Delete an Existing Pronunciation

  1. Select the pronunciation you want to remove.

  2. Select Delete in the window that appears.

  3. Confirm the action to remove the custom pronunciation.


Abbreviation Expansion

This feature allows you to create shortcuts for commonly used phrases. For example, you can create the abbreviation "LOL" for "Laugh Out Loud." When you type "LOL" in the message window, Predictable will expand the text to show the associated phrase, making communication faster and more efficient. Explore the options to add, edit, or delete custom abbreviations.

Add an Abbreviation

  1. Go to "Abbreviation Expansion" in Settings.

  2. Select Add new and enter the abbreviation along with its associated full-form phrase.

Edit an Abbreviation

  1. Go to "Abbreviation Expansion" in Settings.

  2. Select the abbreviation you want to edit.

  3. Select the edit option and make changes to the abbreviation.

Delete an Abbreviation

  1. Go to "Abbreviation Expansion" in Settings.

  2. Select the abbreviation you want to delete.

  3. Select the delete option to remove the abbreviation.