With the evolution of human-like robots, it is no late that Artificial Intelligence(AI) era matches them with real-life voices. The Montreal-based start-up has done exactly the same, world’s first speech synthesis technology skillful of exactly replicating someone’s voice that too with a given emotion was made public by them.
Lyrebird perform the synthesis by analyzing a few seconds of audio recording and then compressing that audio DNA into a unique key. The Al speech generator then produces any speech with its corresponding vocal range. Users can also choose from a wide range of selection from thousands of predefined voices and can also also design a unique voice on the app not faking it.
The synthetic vocal generator developed by Lyrebird is capable of controlling the emotion of generated voices giving them anger, sympathy, stress, and many more human expressions. However, the pinnacle being its ability to produce 1,000 sentences in less than half a second! This puts Lyrebird’s technology in lead of AI speech synthesis.
One of Lyrebird’s co-founder, Alexandre de Brébisson, spoke about how their freshly developed technology requires just about a bit of information to generate like-for-like voices.
“Different voices share a lot of information. After having learned several speakers’ voices, learning a whole new speaker’s voice is much faster. That’s why we don’t need so much data to learn a completely new voice. More data will still definitely help, yet one minute is enough to capture a lot of the voice DNA”.
Also read: What is Identity Theft?
Just like other research studies, Lyrebird’s speech synthesis also faces a few ethical issues like identity theft since only a tiny bit of audio recording is enough to duplicate human voice.
Considering the release of the speech generator, it may be a difficult task controlling millions of voices that will have access to the technology. Moreover, voice recording evidence in criminal cases can also be carried out invalid as it can be of question that the audio file is forged or has been tampered with.
Though, speech synthesis is intended more towards applying this technology in a positive way like providing personal aid for disabled people, the technology also serves an extensive use in animation movies and video game studios.
Lyrebird’s technology is still under development; however, interested individuals can subscribe to the start-up’s website to become a beta-tester or be informed of the launch.