Our story
We started with a voicemail
my teddy talk began with a simple belief: a voice, heard again, can bring more comfort than almost anything else we can hold. What follows is the plain version, including the parts we do not have answers for yet.
On this page
Why we started
In 2022 my dad got sick. It was just after covid. He was ill for about a year, and then he died.
Somewhere in that year he sent me a message. He said he was good, and that he loved me. I have never forgotten it.
“I remember exactly what he said. I have nothing of him saying it.”
I have no recording of that message. No audio, nothing in his voice. Only the words, and my memory of them. That is the whole reason this company exists.
There are other companies that will put a voice inside a teddy bear. You send them a clean recording, they fit it, and the bear plays it back. If you have a recording like that, it works, and you should use one of them.
A lot of us do not have one. And a lot of us never heard the thing we most wanted to hear. Plenty of parents never say it out loud. That is not coldness, it is just how some families are.
So we do it differently. You send whatever you still have of them, however short or rough, and we rebuild the voice from it. Then it says the message you choose.
That is a serious thing to offer and we are careful with it. Some people find a rebuilt voice a comfort. Some find it wrong. Both are reasonable. We say plainly that the audio is a re-creation, we only build one with the family’s permission, and you hear it before anything is made.
It is also not only for people who have lost someone. A good half of what we make goes to families who are simply apart: a parent on a FIFO roster, someone working in another city or another country, a friend you do not see often enough.
Smith, founder
How the voice is made
There are two paths, and they are not the same thing. Which one you are on depends on whether the person can still speak into a microphone.
If they can still record
The recording itself
You, or the person the bear is for, record sixty seconds straight into the site, or send a clean file you already have. That recording is what plays. Nothing is generated, and nothing is changed.
If they have died
A voice rebuilt from what is left
A voicemail, a video, a scratchy few seconds. An AI voice model learns the sound of their voice from that file and reads the message you write. The audio inside the bear is synthesised. It is a re-creation of their voice saying words they never said.
Some people find a rebuilt voice a comfort. Some people find it wrong. Both are reasonable, and you should decide which you are before you upload anything.
What happens to your recording
What the file is used for
Building the voice for your bear, and nothing else. It is never sold, never shared, and never used to train anything.
Where it is processed
A message you record yourself stays in your browser until you order. If you choose AI voice cloning, the sample is sent to our voice provider, because the clone is built on their servers. How we handle voices and recordings.
How long the original is kept
How to have it destroyed
Who builds the rebuilt voice
The full policy
Who is asking
You are being asked to send a stranger the last recording of someone you love. You should know who the stranger is. Right now this site does not say, and rather than fill the gap with a stock portrait and an invented name, here is the gap.
Who founded it
Who makes the bears
Where the studio is
Company details
Compassion first
Every order stands for someone. We treat each recording, message and bear with the care it deserves.
Made by hand
Our makers finish every bear themselves, fitting the audio so it stays safe and clear for years.
Told plainly
We say the word AI where it applies, and we leave a fact blank rather than guess at it. You should know what you are buying.
When you are ready
There is no hurry in any of this. If you have a voice saved somewhere, it will keep until you want to do something with it.
Or ask us something first. A person answers every message.