Profiles & speaking voice
Persona, speaking voice, and appearance
Profiles & speaking voice is where you define and refine the persona layer that keeps your workflow consistent — name, niche, voice, bio, avatar, and speaking-voice settings. Use it to mirror your own voice closely or to design a synthetic persona from scratch.
How to access
Open Settings → Brand profiles, choose Edit profile, then expand Speaking voice. This is the only voice-configuration surface; Create → Speaking voice is for generating and managing finished audio.
Key features
- Profile editing — Update your synfluencer's name, niche, voice description, persona, and bio at any time.
- Appearance & UGC — Set Gender, Ethnicity, Physical Description, Style/Aesthetic, and Preferred Background/Setting. These fields ground UGC photo generation so your synfluencer's visual content stays consistent across images.
- Avatar management — Upload a profile photo or generate one using AI. When you upload a seed image, AI vision auto-extracts appearance details into your profile. You can also generate UGC-style photos for visual content.
- Voice & knowledge — Use its own sidebar page to add an unlimited set of writing, email, documents, transcripts, recordings, and video. Only explicitly approved self-authored passages affect alignment.
- Random talking references — Toggle to allow the video agent to use different visual references for variety.
- Photo management — View, select defaults, and delete generated photos.
- Primary speaking voice explicitly selects one compatible catalog voice or the profile's private authorized clone. Synfluencer uses that saved choice for previews, speaking voice, talking videos, guided recipes, and automations. It never guesses or falls back to a generic voice.
- Cloned voice settings let you record, upload, replace, or preview the private voice assigned to this profile. Creating a clone does not select it automatically; choose it under Primary speaking voice when you want it used.
Creating audio
Open Create → Speaking voice to turn pasted or typed content into a reusable audio file. Synfluencer uses the profile's saved primary speaking voice; if none is saved, follow the Edit profile link and save a compatible catalog voice or private clone before generation. Entering content produces the exact customer estimate; empty content remains an input prompt rather than a pending calculation. Review the final customer estimate and maximum, then start the background task. You can leave the page while it runs. Finished audio appears in a grid where you can play, download, or delete it, and the same file is available under Assets → Audio.