Text to Speech Widget

Use text to speech when you want spoken support without uploading a separate audio file. It is helpful for accessibility, directions, and quick pronunciation support.

Prefer to follow along click by click? Read a question aloud for early readers builds a listening question from a blank lesson: the script, the voice, the recording, and a big, labelled play button.

Where the settings live

Select the widget on the canvas and click the floating ⚙ Configure button to open its setup dialog — the script, the voice, the Generate Audio and Preview buttons, and a live preview of the player all live there. Double-clicking the widget on the canvas opens the same dialog. (Unlike the audio widget, text to speech has no media picker to jump to: it _makes_ its clip rather than choosing one.)

The right sidebar keeps only Style & Layout (colors, button shape and pictures, progress bar, border radius, size and position).

Settings to know

In the ⚙ Configure dialog:

• Text is the script to be spoken, up to 256 characters.
• Voice chooses who reads the script, from Berdee's list of narrator voices: a mix of women's and men's voices, young and grown-up, from bright and playful (good for little ones) to calm and steady (good for older students). Each one shows its name, whether it's a woman's or a man's voice and how old it sounds, and a short description such as "Natural, clear and calm". If you never pick one, the widget uses the lesson's voice, and the picker shows it.
• Writing in another language? Pick Rachel. She reads French, Spanish and other languages natively; the other voices are American English.
• Generate Audio produces the clip. Each clip uses text-to-speech characters from your plan (see AI and Voice). If a clip can't be made — you've used this month's characters, you've made too many clips in a short time, or something went wrong — the reason appears right under the button.
• Preview plays the clip right in the dialog, so you can hear it before you press Save. Press Stop to end it early. Previewing is free — it plays the clip you already made and uses no characters. If the clip doesn't sound right, change the text or voice and press Regenerate Audio.
• Editing the text or changing the voice afterwards does _not_ re-generate the clip. The dialog tells you when the audio no longer matches, and so does the publish check — click Regenerate Audio so learners hear the new wording. You can keep typing while a clip is being made; your edits are kept.
• Auto Play starts playback when the scene loads.
• Loop repeats the generated speech after it ends.

In the sidebar under Style & Layout → Show more options:

• Player Layout controls what students see:
• Button + progress (the default) — a play button and a progress bar in your Progress Color. Students can click or use the arrow keys on the bar to skip around.
• Button + countdown — a play button and the time remaining.
• Button only — just the play button.
• Invisible — nothing is shown and nothing blocks clicks on what's underneath. Use it for speech that plays by itself (Auto Play) or that another widget starts.

The play button is as tall as the widget, so keep the widget at least 44 pixels tall on tablets. A screen reader announces it as "Play" followed by the widget's alt text, so give the alt text a short description like _"the directions"_.

Make the player match your lesson

Under Style & Layout you can restyle the player so it fits the look of your lesson. The settings are grouped by what they change:

• Button: Button Color, Button Shape (circle, rounded square, or square) and Icon Color style the play button. Play picture and Pause picture replace it with your own artwork, for example a speaker character that opens its mouth while it talks. Pick any image from your media library.
• Progress bar (Button + progress layout only): Progress Color, Track Color and Bar Thickness. However thin you make the bar, students can still tap anywhere along its row to skip around.
• Timer (Button + countdown layout only): Timer Color.

A picture is shown whole, on its own, without the colored button behind it or the button shape cut around it, so a transparent PNG looks best. If you set only a play picture, the built-in pause icon appears while the clip plays. If a picture can't load for a student (for example, it was deleted from your library), they see the built-in button instead. Remove a picture to go back to the built-in button.

The picture is drawn inside the button, which is as tall as the widget, so a square image works best. Give the widget alt text that describes what it does, such as _"the directions"_, because a screen reader announces the button by that text, not by the picture.

If a color you pick would be hard for students to see, for example a white icon on a pale yellow button, a warning appears under the settings. Pick a darker or lighter color until it goes away.

Berdee's AI helpers can change a player's colors, shape and bar thickness, but adding button pictures is something you do yourself in the sidebar.

The lesson's voice

Every lesson has one narration voice, and everything that speaks follows it: text to speech widgets and choices that read themselves aloud. Set it once instead of on every widget.

1. In the studio, open the ⋯ menu in the top bar and choose Settings.
2. Pick a voice under Voice. The change applies right away, and Undo puts the old voice back.

A widget follows the lesson's voice unless you picked a different voice for it. Those stay as you set them when the lesson's voice changes, so you can still give one character in a lesson its own voice.

Changing the voice doesn't re-record anything by itself. If the lesson already has audio in the old voice, Settings tells you how many clips are out of date, for example _"3 text to speech clips and 4 read-aloud clips are in another voice or out of date"_, and offers Regenerate audio, which makes them all again in the new voice. Each clip uses text-to-speech characters from your plan, as Generate Audio does. Until you regenerate, students hear the old clips (text to speech) or the device's voice (choices), and the publish check reminds you.

A lesson made before this setting existed keeps the voice its widgets already use: the voice most of its text to speech widgets use, and Jessica if it has none.

One voice at a time

Students hear one voice at a time. When a clip starts playing, whatever else was speaking stops: another audio or text to speech widget, a choice reading itself aloud, or the device voice counting in a counting collection. And when one of those starts, this clip pauses. So a child who taps a choice while your instruction is still playing hears the choice, not both at once. Press play to pick the instruction up where it stopped.

Loop is the exception: looping audio is background music, so it keeps playing under every voice and nothing stops it.

Before you publish

• No audio yet. On the canvas, a text to speech widget with no clip says No audio yet, and students would see "Audio isn't available yet" instead of a player. The publish check lists it too.
• Audio is out of date. The text or voice changed after the clip was made.
• Can't be stopped. An Invisible player with Auto Play and Loop both on speaks forever with no way for a student to stop it. The dialog and the publish check both warn you — turn off Loop or Auto Play, or pick a visible layout.

These are warnings: you can still publish, but students will get exactly what the warning describes.

Grading setup

Text to speech widgets are not graded directly. They support access, pacing, and comprehension.

How to set up the content

Keep the spoken text concise and purposeful. This widget is strongest for short directions, modeled pronunciation, and short passages that benefit from audio access.

Useful setting combinations

• Short text + visible player:

Best for repeatable learner support.

• Short direction + Auto Play on:

Best when the scene should speak immediately on arrival.

• Pronunciation cue + Loop off:

Best when the learner should replay only when needed.

• Generated speech + hidden feedback text:

Best when listening should unlock a follow-up question or explanation.

Strong use cases

• Read-aloud support for key directions.
• Pronunciation support in language learning.
• Alternate access for emerging readers.

Great pairings with other widgets

• Pair TTS with text so learners can read and hear the same content.
• Pair it with text input for dictation or echo writing. Give a spelling test students listen to builds a three-word test this way.
• Pair it with multiple choice for listening checks.
• Pair it with flashcard so the learner can hear the front or back content.