Less than a year has passed since we launched the Voice Marketplace. Respeecher has given users access to technology that, until recently, was only available to large video game and motion picture studios. If you haven't heard of the marketplace yet, we recommend reading our initial announcement.
In short, the Voice Marketplace allows you to license a human voice from Respeecher’s voice library. Next, you can synthesize an unlimited amount of speech using this voice for your project, within the same platform, leveraging the capabilities of an AI voice generator. All this at an amazing price point and quality that no one else can match.
Here’s a bit on the tech side
We have dramatically reduced the learning curve for our ML model. Calibrating the system now requires less than five minutes of the source recording before you can begin synthesizing speech.
This was made possible by our engineers employing the latest innovations in deep learning - transfer learning and advanced self-supervised learning.
Although, we often encounter challenges when working with old recordings because some of them suffer from quality issues of all kinds. But thanks to constant improvements to our voice AI algorithms, we deal with all the challenges and deliver the highest resolution audio across the board. You can find out more by downloading this whitepaper about Respeecher's audio super-resolution algorithm.
If you're the kind of nerd who wants to know everything, this is how it works.
We’re using a part of our neural network that has been trained using thousands of hours of speech to generate universal phonetic representations that are valid for any possible speaker.
Another part of the network focuses on working with those features of the voice that are unique to the source speaker.
Thus, we were able to reduce the time it takes to create a working model for your unique voice from weeks to minutes. How? By employing features of a human voice that are pretty much common to all of us, through AI voice synthesis.
So what exactly has changed in the client experience?
Below are the steps of the Voice Marketplace’s updated onboarding process:
-
Sign up for a free trial on the Marketplace registration page.
-
Next, you'll need to upload a sample of your voice (about 5-minutes long) to calibrate Respeecher.
-
Pick up any voice from our library and synthesize an unlimited number of 30-second samples while on your free trial.
In short, from the moment you landed on our site to the moment you began synthesizing your first minutes of speech, only took about 5 minutes for registration + calibration. Whereas in the past, this entire process would have easily taken more than two weeks.
Wrapping up
Today, the Voice Marketplace offers unique opportunities to small studios and content creators, and in doing so, has significantly democratized the synthetic media market.
Beyond making the service affordable for production, consider that you gain unlimited possibilities for its use by licensing a synthetic voice. And this is a major advantage over hiring and licensing the services of real voice actors.
When you use Respeecher to license a voice, you are no longer dependent on that actor's schedule or availability. The synthetic voice will stay with you for as long as you need, allowing you to create a character, find the most suitable voice, replace your voice if you don't like it. But what about creating a new voice, you may ask? Respeecher can do that too and you can find all the answers to questions about speech-to-speech voice conversion on our FAQ page.
Voice cloning makes audio post-production a breeze. Just browse our blog for more use cases and benefits of speech to speech technology.
In the meantime, we are happy to speak with you about your project and answer any questions you might have about possible cooperation. Tell us about yourself today and we'll find out how Respeecher, with its advanced AI voice generator capabilities, can be of use.
