We are excited to announce the latest version of Revoiceit, which features improved functionality to improve specific use cases and help customers achieve their audiovisual localisation goals. This update introduces significant advances in audio quality, synchronization accuracy and language functions, consolidating Revoiceit as a powerful tool for AI dubbing and production.
With its new features, Revoiceit not only improves the efficiency and accuracy of the virtual voice production process, but also now offers an even more intuitive and user-friendly experience, allowing users to easily create their own content and localize high-quality AI voices through a fully customized process.
Enhanced Audio Quality
First of all, Revoiceit v2.3 introduces new speech models with significantly improved audio quality and overall prosody. These improvements ensure a more natural and expressive voice production, increasing the realism and expressive impact of the generated audio.
With these new models, superior speech synthesis is achieved, suitable for a wide range of different use cases (e-learning, educational content, training, podcast, advertising, social media and much more).
Revoiceit now also offers users three different prosodies for each line upon initial generation. This provides flexibility and choice of sentence prosody, allowing users to listen to variations and select from the most suitable take for their project.
Enhanced Synchronization Accuracy
Revoiceit v2.3 introduces several new features aimed at improving audio generation and synchronization:
> Timeline Synchronization in the Production workspace
For content without source audio, it is now possible to upload video and then synchronize the platform-generated audio directly within the Production workspace (this was previously only available for Dubbing workspaces). This functionality is particularly useful for instructional and e-learning videos that do not have a source voice to be dubbed; users can now create AI-generated voices from scratch.
Direct Timeline Editing, including:
> Timeline Zoom-in
To enhance precision, users can now ‘zoom-in’ on the timeline, offering a detailed view of the audio waveform. This allows for meticulous cutting and adjusting, ensuring perfect synchronization.
> Fade In and Fade Out for Audio Chunks
To improve synchronization adaptation and help reduce noise, chunks that are cut within a segment now get a fade-in/fade-out effect, providing a smoother audio experience.
> Manual Editing of Start and Stop Points
In the source assets review page, users can manually edit segment start and stop points. This precise segmentation of source text ensures a customized alignment with the target language, improving the overall dubbing accuracy.
Advanced Linguistic Features
The latest update also brings also new linguistic features to enhance the pronunciation setting workflow:
> Glossary Additions from Preview
Users can now add new glossary terms directly from the Text Editor in the Preview Page, streamlining the workflow and ensuring consistent terminology.
> Editing Existing Terms in the Glossary
This feature allows for easy updates and corrections to the glossary, ensuring accurate and up-to-date term usage.
Expanded Language Support
We are pleased to announce that with this release, Revoiceit increases its language coverage, now supporting Simplified Chinese, Japanese, Korean, Turkish, Hindi, and Portuguese (Portugal), and reaching the following fifteen output languages:
- US English
- UK English
- French
- Italian
- German
- Spanish (Spain)
- Spanish (LatAm)
- Portuguese (Brazil)
- Portuguese (Portugal)
- Polish
- Simplified Chinese
- Japanese
- Korean
- Turkish
- Hindi
This expansion allows more users to leverage Revoiceit’s capabilities for their localization needs, making it an even more versatile and powerful tool.
Stay tuned for more updates as we continue to innovate and enhance Revoiceit - the professional virtual studio for expressive AI voices.