EMOTomo lets you speak to characters and hear them speak back. This notice explains exactly what happens to audio. It supplements our Privacy Policy.
Voice input is push-to-talk: the microphone is used only when you tap the mic button. EMOTomo does not listen in the background and does not access your microphone when a page simply loads. Your browser will ask for microphone permission, which you can grant or deny.
When you record, your audio is sent to a speech-to-text service to produce a text transcript. By default this is done by our own self-hosted speech model. As a fallback, transcription may be performed by ElevenLabs. The resulting transcript becomes a chat message only if you choose to send it.
Your microphone audio is used to create the transcript and is then discarded. We do not save raw voice recordings, and we do not build voiceprints or use voice for biometric identification.
Once your speech is a text message (or if you type), the text is sent to the AI model provider that powers the character's reply, along with the conversation context. See our Third-Party Processors list.
Character replies can be turned into speech using ElevenLabs text-to-speech. The character's reply text is sent to the provider, and the generated audio is briefly cached so it can play back to you, then expires.
You can use EMOTomo entirely by text and never enable the microphone. You can revoke microphone permission in your browser at any time.
Questions: admin@emotomoai.com. For data requests, see our Data Deletion Policy.