Overview of Supertone: Advanced Voice Synthesis and Conversion Technology

Supertone is a media products company specializing in intelligent audio technologies for various creative and commercial applications. The company's offerings include a range of tools and systems designed to enhance and transform audio content through advanced AI-driven processes.

Key Features and Products

Project Screenplay

Functionality: An authoring tool that allows creators to write stories based on scripts and provide them with hyper-realistic voices using AI technology.
Applications: Suitable for content ranging from short scripts for podcasts and YouTube videos to full-length game scenarios.
Capabilities: Includes writing, editing detailed performances, expressions, and direct voice guidance for enhanced detail.
Collaboration and Export: Features easy export options and collaborative sharing functionalities.

Supertone Shift

Description: A desktop application designed for real-time voice conversion with minimal latency and high audio quality.
Unique Selling Point: Offers advanced real-time voice conversion suitable even for live performances.
Customization: Users can modify their voice while maintaining their unique identity.

Supertone Clear

Purpose: An audio plugin aimed at enhancing dialogue clarity for video and audio content creators.
Features: Allows control over voice, reverb, and noise levels with three intuitive knobs.
Technology: Utilizes a highly efficient neural network to balance sound in real-time, suitable for professional studio work.

Research and Development

Areas of Focus:
- Singing voice synthesis
- Text-to-speech synthesis
- Grapheme-to-phoneme conversion
- Melody and lyrics transcription
- Studio-quality speech enhancement
- Voice conversion
- Natural language processing
- Speaker verification
- Automatic speech recognition

Publications & Awards

CES 2022: Innovation Awards Honoree in Software & Mobile Apps.
NeurIPS 2021: Featured publication on reconstructing speech from self-supervised representations.
ICASSP 2021: Achievements include 3rd place in the real-time denoising track and publications on denoising and dereverberation.

Use Cases

Music

Capabilities: Creation of entirely new voices and control over voice characteristics for songs and live performances.

Video Production

Features: Voice alteration tools (age, gender, diction) and natural multi-language dubbing for global distribution.

Gaming

Innovations: Face-to-voice matching AI technology for character voice design and dubbing, enhancing global release schedules.

Corporate Identity

Services: Custom voice creation that embodies a brand’s identity, with technical support for voice activation.

Ethical Considerations

Supertone is committed to ethical AI practices, ensuring that:

Voice monetization occurs only with the owner's permission.
Non-commercial research is limited to public or deceased figures.
Data privacy is maintained with restricted access to training and synthesized voice data.
Creative culture is respected, enabling a wide range of content creation while addressing personal information concerns.

Collaboration Process

Interested parties can engage with Supertone through a structured process that includes a business proposal review, agreement signing, content production, and launch preparation, ensuring a minimum production timeline of three months.

For more information or to explore potential collaborations, visit the Supertone product website.