Not so long ago, high-quality dubbing was only available to large animation and film studios. Due to the high costs of hiring voice actors and sound engineers, most content producers could not afford to produce professional dubbing in a required variety of voices.
Beyond that, the art of dubbing demands weeks’ worth of time and care during post-production. Additional dialogue replacement alone can drag on for weeks and diminish the morale of any small team.
Thanks to artificial intelligence and machine learning, Respeecher has developed an AI voice technology that allows for the transformation of any voice into the target speech of any person.
We have revolutionized the idea of accelerating the dubbing process while drastically optimizing the logistics and timelines associated with it. Respeecher is going one step further in democratizing the voice market by making AI voices available to small content creators, including one-person studios.
Respeecher Voice Marketplace in a nutshell
Imagine you are a small studio working on an animated film. You have a staff of five and every team member is passionate about their work. Every penny counts.
You can't afford to bring in ten professional voice actors to dub every character in your upcoming movie. Perhaps you can afford five. But what if we told you that you only need one actor, and all your characters will speak with unique voices (no matter their age or gender)?
How is that possible, you ask?
- You only need one professional voice actor. Strictly speaking, you could even do it yourself. But still, it is always better if the original dialogues are spoken by a professional. This person dubs all the dialogues in your piece. We then use their speech (records) as a source to generate unique voices for your movie’s characters. With their consent, of course.
- This is where Respeecher's Voice Marketplace comes into play. You can choose any voice from a variety of virtual personas generated by us. Each AI voice is unique, and the rights to use it are much easier to transfer to you than to transfer the rights of a real-life actor.
- After you have chosen unique voice for each character in your film, we will ask you to provide us with the recordings of the input (a source) voice actor. Our system will analyze the actor's speech and transform it into the AI voices you have licensed from us.
- As a result, you get ten soundtracks performed with unique voices, preserving all the intonations and moods of the source voice.
You can find out more by visiting the Voice Marketplace on our website. Currently, the marketplace is in an early launch phase. We provide everyone with a free 7-day trial so that they can experience all the benefits for themselves. Please don’t hesitate to contact us with any questions or proposals.
Working to make the Voice Marketplace even better
As we’ve just described a typical AI voice generation workflow you can expect when using the Voice Marketplace, now we’d like to introduce you to the main elements that we are improving or planning to add to the platform in the near future:
- The ability to convert long recordings (not limited to 15 seconds).
- Improving the quality of by reducing quantization noise.
- Improving the overall speed of voice conversion.
- Introducing the 44khz model instead of 22khz (improved sampling rate).
- Reducing data requirements from a single user, eventually getting rid of requiring any data from a user and therefore pre-training. We want you eventually to be able to use it right away, the same day.
- Improving overall quality by making generated speech sound perfectly from the first and only take.
- Introducing dozens of new voices.
- Adding more languages to the platform (Spanish, Brazilian, Portuguese, Chinese, Hindi, German, and Italian are among the first on the list).
Over the next six months, we’re committed to gradually improving the tool. Taking into account all the first users' feedback we plan to complete preparations for a final release.
By this time, we will be working with hundreds of clients around the world, providing them with access to dozens of unique voices available in multiple languages. Stay tuned and subscribe to voice synthesis updates below. We're just getting started!
