Update Logs
View product update logs, including latest features and improvements.
Talking Avatar and Workspace Polish
This update connected finished voices to more visual output and tightened the workspace experience.
What changed
- Renamed Lip Sync to Talking Avatar across routes and workspace navigation.
- Added talking avatar examples for cinematic, game, and cartoon animation use cases.
- Improved My Works so generated voice and video results stay inside the workspace shell.
- Unified workspace UI colors around black, white, and RoleTTS green.
- Improved voice browser loading so more voices load automatically when scrolling.
- Replaced old demo blog content with RoleTTS product content.
Why it matters
RoleTTS now feels more like one connected creative workspace: create a voice, save it, review your work, and carry the result into talking avatar output.
Voice Clone with Before-and-After Review
This update focused on making cloned voices easier to judge and reuse.
What changed
- Added upload and record modes for creating cloned voices from short audio.
- Added a cloned voice preview section with playback, download, and save-to-library actions.
- Simplified the clone page into separate cards for input, preview, and saved voices.
- Tuned the before-and-after messaging around tone, pacing, and vocal texture consistency.
Why it matters
Voice cloning is only useful when the result stays close to the original reference. This release made that comparison clearer before a voice is saved.
Voice Design from Image and Prompt
Voice Design introduced a more visual way to create character voices.
What changed
- Added a Voice Design workspace for creating voices from a reference image and prompt.
- Added generated voice preview so creators can review the result before saving.
- Added My Designed Voices for keeping designed voices available later.
- Improved example cards to show how image mood and prompt direction can shape a voice.
Why it matters
Instead of only searching for an existing voice, creators can now describe and visually guide the voice they want.
Expressive Voices and Directed TTS
This update made voice selection faster and more useful for character content.
What changed
- Expanded the preset library around expressive character voices.
- Added cleaner voice browsing with language, gender, age, accent, and category filters.
- Improved the Text to Speech layout so script, voice, model, and generation controls stay connected.
- Added better voice preview behavior for browsing and auditioning voices before use.
Why it matters
Creators can now start from a larger library of voices and quickly narrow the result to the role they are casting.
RoleTTS Workspace Foundation
RoleTTS started with a focused workspace for creators who need directed voice output without jumping between tools.
What changed
- Added the core Text to Speech workspace for writing scripts and generating audio.
- Added the first voice selection flow with reusable voice profile state.
- Added a voice library foundation for browsing preset character voices.
- Added saved voice state so selected voices can carry across the workspace.
Why it matters
This release established the baseline: write a line, choose a voice, and generate audio from the same surface.