Antiphon: Voice Preview
Everyone
info
0+
Downloads
Everyone
Learn more
About this app
Antiphon is a limited voice-cloning preview for Android. Record your own voice, or a voice you have explicit permission to clone, and hear it speak eight selected English phrases. Guided recording, saved voice profiles and repeatable phrase playback help you explore how a recording affects your clone. Bundled demonstration voices let you try playback before creating a profile.
The voice models download during setup. Recording and voice generation then run on your device. Generation time and voice similarity vary with your phone and recording; a clone may not closely match its speaker. Generated speech can mispronounce or change words. Listen to each clip before using it.
DISCLOSED VOICE PREVIEWS
Every new preview begins with a fixed computer voice saying "This is synthetic speech made by a computer." Your selected voice then speaks the preview phrase. Listen before sharing and confirm that you clearly heard the disclosure. Shared preview files also carry an AI-generated filename, caption and WAV metadata. Use voices only with permission and make clear to listeners that the audio is synthetic. The permission declaration is not independent identity verification.
Replay recent generated clips in History and share new preview clips after reviewing the disclosure. Report an offensive or misleading result from inside the app. Reporting requires a network connection.
FIRST-EDITION LIMITS
This edition offers selected English phrases only. Freeform text, translation, Listen, audio-file transcription, Chat, Soundboard and Full reference mode are unavailable. Older unrestricted generated clips remain available for playback but cannot be shared from this edition. This update preserves your recordings and existing History; the app's existing rolling History limits still apply.
LARGE DOWNLOADS
Android 8.0 or newer and a 64-bit ARM device are required. Initial speaking files total about 1.48 GB; creating your own voice needs about 0.61 GB more. Use Wi-Fi and allow at least 3 GB of free storage. Models download separately, resume after interruption and are checked before use. Devices with limited memory may generate slowly or be unable to complete a job.
YOUR DATA
There is no account, advertising or in-app purchase. Recordings, voice profiles and generated audio stay on your device; generation does not upload its input or output. If you explicitly send a content report, it sends your category, description, app version, platform and random receipt ID over HTTPS. Generated text is included only if you separately select it; audio and voice profiles are never attached. Reports are kept privately for up to 30 days. Contact support with your receipt ID to request earlier deletion.
The included Google ML Kit SDK collects diagnostic, usage and device or installation identifier information. Model hosts receive ordinary network request information. Read the privacy policy for details, including SDK retention. Local app data is excluded from Android cloud backup and device transfer. Clear app storage or uninstall to remove it; delete copies you exported separately.
The voice models download during setup. Recording and voice generation then run on your device. Generation time and voice similarity vary with your phone and recording; a clone may not closely match its speaker. Generated speech can mispronounce or change words. Listen to each clip before using it.
DISCLOSED VOICE PREVIEWS
Every new preview begins with a fixed computer voice saying "This is synthetic speech made by a computer." Your selected voice then speaks the preview phrase. Listen before sharing and confirm that you clearly heard the disclosure. Shared preview files also carry an AI-generated filename, caption and WAV metadata. Use voices only with permission and make clear to listeners that the audio is synthetic. The permission declaration is not independent identity verification.
Replay recent generated clips in History and share new preview clips after reviewing the disclosure. Report an offensive or misleading result from inside the app. Reporting requires a network connection.
FIRST-EDITION LIMITS
This edition offers selected English phrases only. Freeform text, translation, Listen, audio-file transcription, Chat, Soundboard and Full reference mode are unavailable. Older unrestricted generated clips remain available for playback but cannot be shared from this edition. This update preserves your recordings and existing History; the app's existing rolling History limits still apply.
LARGE DOWNLOADS
Android 8.0 or newer and a 64-bit ARM device are required. Initial speaking files total about 1.48 GB; creating your own voice needs about 0.61 GB more. Use Wi-Fi and allow at least 3 GB of free storage. Models download separately, resume after interruption and are checked before use. Devices with limited memory may generate slowly or be unable to complete a job.
YOUR DATA
There is no account, advertising or in-app purchase. Recordings, voice profiles and generated audio stay on your device; generation does not upload its input or output. If you explicitly send a content report, it sends your category, description, app version, platform and random receipt ID over HTTPS. Generated text is included only if you separately select it; audio and voice profiles are never attached. Reports are kept privately for up to 30 days. Contact support with your receipt ID to request earlier deletion.
The included Google ML Kit SDK collects diagnostic, usage and device or installation identifier information. Model hosts receive ordinary network request information. Read the privacy policy for details, including SDK retention. Local app data is excluded from Android cloud backup and device transfer. Clear app storage or uninstall to remove it; delete copies you exported separately.
Create a voice clone and hear selected English preview phrases on your phone.
Updated on
Safety starts with understanding how developers collect and share your data. Data privacy and security practices may vary based on your use, region, and age. The developer provided this information and may update it over time.
- No data shared with third partiesLearn more about how developers declare sharing
- This app may collect these data typesApp activity, App info and performance, and Device or other IDs
- Data is encrypted in transit
- You can request that data be deleted
What’s new
Keep using your saved voice profiles after updating.
Choose output languages by their full names.
Start faster with a clearer Standard generation option.
Get step-by-step Help with highlighted controls.
Choose a portrait guide and preview its bundled voice. Spoken Help uses on-device voices and remains in English for now.
Choose output languages by their full names.
Start faster with a clearer Standard generation option.
Get step-by-step Help with highlighted controls.
Choose a portrait guide and preview its bundled voice. Spoken Help uses on-device voices and remains in English for now.
Everyone
Learn moreApp support
About the developer
MICHAEL ANOBILE
contact@aethericresonance.dev
Canada
