iPhone · On-device models
Aftertone
Record. Write. Change the voice.
Listen to the transformation
Recording conversion
Ironbeard
Recorded samples
Recorded comparison
Speech from text
Give a written line a voice.
Hear speech from text
THE WRITTEN SENTENCE
“Please leave the package by the kitchen door. I will call you after lunch.”
The same written sentence, spoken in the selected voice. These are recorded outputs, not live generation.
Recorded development outputs. All four voices are original synthetic characters. Processing time is not represented.
Give a written line a character voice.
Use a recording for voice conversion, or start with a written line for speech. Listen to the selected result before sharing from the app.
For character-voice recordings on iPhone. Models run on your device in an offline workflow; downloads and device requirements vary by feature.
Aftertone details
Who it’s for
For turning a recording or a written line into a character voice.
What you can do.
Convert a recording.
Record or import audio, choose a reference voice, then compare the conversion with your original and share the result.
Create speech from text.
Type a line and choose a reference voice. A separate on-device engine creates speech you can listen to and share.
Before you start.
Availability
Release details pending
Store access, pricing and release dates will be added when confirmed.
- On your iPhone
- Recording conversion and speech from text use models running on your device.
- Feature requirements
- Different features have different iOS, memory, model-download and speech-asset requirements. Final compatibility is pending.
- Recorded audio
- Voice conversion is an offline workflow. It does not change live calls.
Recorded evidence
Same recording. Another voice.
Recorded development outputs · Original synthetic character voice. Processing time is not represented.
Original recording → Ironbeard
Experimental delivery controls
Captured development interface. Delivery direction is experimental and results are not guaranteed.
View the full app screenWhat this preview shows
The recording, converted voices, and text-to-speech samples are actual development outputs. The full audition features four original synthetic character voices. Processing time is not represented. Delivery-direction controls in the development interface remain experimental.