Google is launching a new AI voice synthesizer which allows developers or businesses use the text-to-speech synthesis that powers the voices in Google Assistant and Maps.
According to Google’s blog post, the new service can be used to bring advanced artificial voices to a variety of areas, such as voice response systems for call centers, conversations with IoT devices, and converting text-based media to audio. There are 32 basic voices to choose from, across 12 languages and it also allows a user to customize pitch, speaking rate and volume gain. Some languages even have a range of male and female voices available.
The service was developed by Google’s DeepMind team, which has been used to make the Google Assistant sound more natural. Rather than using fragments of speech and stringing them together to make words — which often sounds very robotic, the updated version of WaveNet forms individual sound waves, creating more natural sounding speech. This is said to be 1000 times faster than before, fidelity increased to 24,000 samples per second and the resolution has been bumped up from 8 to 16 bits, all of which reduce the quality gap between generated speech and human speech by about 70 percent according to tests done.