What to record for message banking

Message banking means recording phrases in your own voice to play later on a communication device.

Message banking was developed at Boston Children's Hospital as a way for people with ALS (MND) to keep recordings of their own voice, with their own emotion, for use later.1 A recorded message can't say anything new,2 so many people also use a synthetic voice.

Ideas for what to record

Names and greetings

  • The names of the people you talk to most, the way you say them, including nicknames.
  • How you say hello and goodbye: "Hi, love." "Morning, mate." "See you later, alligator."
  • How you answer the phone.

Things you say to family and friends

  • "I love you." "I'm proud of you." "Love you loads."
  • Goodnight routines and bedtime phrases. The MND Association suggests reading a bedtime story as an example.3
  • Encouragement: "You've got this." "Good luck today."

Familiar sayings and sounds

  • Catchphrases, sayings and jokes only your family understands.
  • Reactions: "You're kidding!" "Oh, for goodness' sake." "Brilliant."
  • Sounds that aren't words: a laugh, a sigh, a "hmm", a whistle.3
  • How you call the dog or the cat.3

Everyday needs

  • "Can you help me with this?" "I'm fine, just tired." "Not now, thanks."
  • Things you'll want to say to carers or nurses, in your own voice.

Things for later

  • Messages for birthdays or special occasions.
  • Favourite prayers, songs, poems or lines from films.

Tips for good recordings

  • Pick a good timeRecord when you're rested. If your voice tires through the day, mornings are often best.
  • Find a quiet spotA room with curtains, carpet or soft furniture, away from fridges, fans and televisions.
  • Hold the phone steadyAbout a hand's width from your mouth, a little to one side so breath doesn't pop into the microphone.
  • Say it naturallyIt can help to imagine saying it to the person.
  • Listen backCheck each take before moving on. Redo any with background noise.
  • Label clearlyName each recording so you can find it fast later.
  • Back upKeep a copy somewhere other than the phone.

How to record in Dawn

  1. Open the Phrases tab and tap Record.
  2. Say the phrase aloud in your usual voice.
  3. Dawn saves it as a button on the Quick Phrases board, marked with a small waveform. Rename it so it's easy to find.
  4. Make a backup in Settings, then Data & Backup. Recordings are included.
Dawn's Message banking screen with a record button and the recorded phrase 'Love you loads'
Each recording becomes a button on the Quick Phrases board.

For messages you haven't recorded

Dawn can also build a synthetic voice from your speech for anything you type. Making a voice in Dawn.

Questions people ask

How many messages do I need?

There's no target. Some people start with a dozen and add more over time.

Should I record single words or whole phrases?

Mostly whole phrases, said the way you'd normally say them, so they keep your timing and expression. Single words recorded alone tend to lose that. A synthetic voice can cover anything you haven't recorded.

Can I record things for a specific person, like a child's birthday?

Yes. You can record a message for a birthday or another future occasion.

What if I don't like how I sound?

Listen back and record another take if you want to change it.

Sources

  1. Costello J, Smith M. The BCH message banking process, voice banking, and double-dipping. Augmentative and Alternative Communication, 2021. doi.org/10.1080/07434618.2021.2021554
  2. Pu S, et al. Exploring voice banking as an alternative augmentative communication strategy for individuals with dysphonia, aphonia, and dysarthria: a scoping review. Journal of Voice, 2025. doi.org/10.1016/j.jvoice.2025.10.018
  3. MND Association. Information sheet 7D: Voice banking and message banking, January 2026 (PDF). www.mndassociation.org/sites/default/files/public/2026-01/7D-Voice-banking-and-message-banking.pdf

Try Dawn on iPhone or iPad

Symbols, typing and saved phrases, spoken in a natural voice. Dawn can also build a voice from about eight seconds of your own speech.

Requires iOS 17 or later.