NExt-Gen Vocal Synthesis


About emvoice


Creating a New Vocal Experience to Power the Coming "Ai-First" World

Speech synthesis is one of the

“key blockers to truly enjoyable and natural Voice Interfaces”

according to J. Zimmerman who is the Google Conversation Experience Design Lead.

We aim to solve this problem with EVE (Emvoice Vocal Engine), a revolutionary text-to-voice (TTV) speech technology.

Expressive singing and speaking

We’re building a tool that can sing, speak, and smoothly transition between these two ways of expression, just like humans. (And in case you’re wondering, rapping will be covered too.)

 We have seen how vocal assistants are rapidly improving and learning social skills. Vocal companions, social robots, smart messaging apps, augmented/virtual reality: these markets are growing extremely fast or quickly emerging. And they need highly enhanced vocal capabilities to deliver on their extraordinary promises. This is Emvoice's call-to-action!

EVE, A Breakthrough Technology

EVE (Emvoice Vocal Engine) can not only sing, but also speak in a much more expressive way than any other TTS (text-to-speech) technology. Expression, when speaking and singing, is based on continuous and sometimes drastic changes in three parameters: the pitch envelope, the length of the syllables, and the use of various vocal modes. Current TTS engines are unable to take these parameters into account, and whatever little “expression” they have comes from the original recording of the voice. EVE works the opposite way. It removes the expression of the original recording and then breathes new life into it. 





Soho is the consumer version of our engine and is targeted at music producers and creators. Soho is a virtual instrument plugin for producers to allow them to add synthetic vocals to their productions for the first time ever.

Successful music is almost exclusively vocal. But it's difficult, expensive, time-consuming and often impossible to record vocals, and most music producers can't use live vocals. Soho will change that. 

Soho will also be the testbed for our core technology.


Emvoice Vocal Engine (EVE)


EVE is a uniquely capable text-to-voice engine that can:

  • Convey meaning, express emotions, and make conversations engaging
  • Generate any sentence in an almost unlimited number of ways 
  • Adapt to differing interactions like "fun, high-quality conversation", reading the news, telling a bedtime story, singing a duet or reading poetry

EVE has the potential to be the voice of the rapidly growing number of conversational UI's.

We're tapping into a new market with billions of daily users. 20% of Google queries already use voice, our most efficient, and most natural way to communicate. 



the emvoice TEAM


Rodolphe OLlivier

CEO, founder, recording studio owner. Rodolphe is the architect of Emvoice.


David M. Papurt

CTO, international software consultant and architect, holds a Ph.D. from MIT, a BA in Music and five US patents.


Olivier Mohsen

Lead Developer, extensive knowledge of audio, coding and music. Genius-level mathematician.

Richard moon

Head of Product, holds four design patents and was Head of UX at Bose and Vodafone. 


Next Steps


See our jobs listings on AngelList.

Vocal Talent

We are looking for disciplined and flexible individuals to be the singing and talking voices of the future; get your voice featured in thousands of productions and be the "voice in the machine". See details here.


Come join us. We are looking to raise our seed round from engaged, industry investors.

Contact us below to get an investor pack.

Emvoice is backed by Techgrind, Southeast Asia's first private Technology Incubator.

ProduceRS & music creators

Soho, our first commercial product is coming soon. Watch this space! See our demos here or our vocal examples on soundcloud. Contact us below. We promise, no spam, ever!


get in touch