Introducing Resemble Clone – a creative tool for crafting speech

We are delighted to announce the launch of Resemble Clone, designed to help entertainment industry creatives synthetically produce high-quality voices. Resemble Clone enables users to record a few sentences in their own voice and immediately generate high-quality samples. With just three minutes of audio data, Resemble Clone can create the audio track for an immersive VR experience, an animated feature film, generate an entire audiobook, or power the voice of a one-of-a-kind Alexa Skill. Resemble AI solves the large scalability problems that creatives face when creating speech content. We are excited to have Craft Ventures, Betaworks Ventures, AET Fund, and firstminute Capital as partners on this journey.

We set out to build a product that helps creatives get over the hurdle of crafting audio content. With more audio content being produced year after year – smart speakers, Airpods, podcasts, audiobooks, and digital characters in virtual and augmented reality – there is a large and growing need for fast and accurate voice cloning. Resemble AI’s unique focus is to empower creatives, so they can control and produce content without sacrificing quality.

“The team at Resemble AI is doing something special. They are a rare combination of thoughtful and capable. Their deep understanding of the underlying engineering behind synthetic media, in particular how novel methods can generate and replicate true-to-life speech patterns, is matched with a practical bearing to make these powerful technologies accessible for wide-scale distribution in a responsible manner,” said Sankalpa Ghose, CEO of Anart AI. “Whether for industry, research, or creative pursuits, they are the group to partner with – both now, as reliably updated platform for content-first productions to come.”

How Resemble Clone Works

• A powerful web-based editor allows users to tune the expressiveness of the speech content down to the sound byte. 
• Proprietary algorithms smooth out the audio to make it sound realistic.
• The long, tedious, and linear process of creating new speech content is eliminated. 
• Users can start creating and consuming new content within seconds. 
• Resemble’s API makes handoffs to developers simple. 

“Resemble has built an easy-to-use, self-service platform that will enable the next wave of content generation via synthetic voice,” said Bryan Rosenblatt, principal investor at Craft. “Their ability to translate such a complex technology into an easy-to-use product is a creative enabler that we’re excited to support. We are also glad to see their continued investment in deep fake detection, ensuring this technology is used appropriately.”

“We’re excited by the idea of Resemble making real-time creation and editing of audio content – which today is a painful bottleneck for creatives across industries – as easy and accessible as editing animated visual content,” said Clara Lindh Bergendorff, investor at firstminute. “Resemble is also well-positioned to ride wider audio waves, from the growth of audio content consumption and voice applications (e.g. voice shopping expected to reach $40bn in the next three years), to growth in audio-first devises (e.g. Amazon’s Alexa expected to hit 200m units this year, and Apple’s airpods expected to generate $20bn in revenue in 2020).”

Protecting your Voice

We are aware of the challenges that come with new generative technology and have heavily invested in making sure all voices are protected at all costs.

Resemblyzer – an open-sourced project developed by the our team – uses AI and Deep Learning to detect deep fakes. For those that want to protect against fake voices in the wild (outside of Resemble), Resemblyzer uses a deep learning model that derives a high-level representation of voice samples and is able to predict whether they are real or generated.

As researchers and entrepreneurs, we are thoughtful about the benefits and/or risks to society of what we are creating. When you’re creating your voice on our platform, we take extreme measures to ensure the ownership of the voice.

About Resemble AI

Resemble AI is creating synthetic speech that is focused on human emotion. With state of the art Artificial Intelligence, Resemble Clone is the first step to enabling creatives to craft speech content with as much freedom and flexibility that the graphics world enjoys today. Founded in 2019, Resemble AI is headquartered in Toronto, Ontario. To learn more, please visit https://www.resemble.ai.