Microsoft Azure Intelligent speech Synthesis is fully upgraded to 48kHz High Fidelity Model

CTOnews.com, November 17 (Xinhua) Microsoft Azure neural network text-to-voice service (also known as "Neural TTS", "intelligent speech synthesis") can help users convert text into realistic artificial intelligence voice, which is suitable for a variety of application scenarios, including intelligent voice assistant, customer service conversation robot, audio content reading, game character voice, and so on. In the past few months, Microsoft Azure intelligent speech synthesis technology has made rapid progress in speech naturalness, sound richness and multilingual support.

Today Microsoft officially brings you the latest neural network speech synthesis vocoder HiFiNet2.

Vocoder is one of the key components in TTS, which synthesizes audio samples based on input text or acoustic features. At present, Microsoft has upgraded its Azure intelligent speech synthesis products to 48kHz sound model through HiFiNet2 vocoder technology, which further brings users a more fidelity, efficient and scalable AI voice quality experience. The update includes more than 400 tones and covers languages in more than 140 countries and regions around the world.

48kHz speech model

In the text-to-speech technology, audio fidelity is an important standard to measure sound quality. High-fidelity sound can not only convey richer and more delicate sound quality to users, but also minimize the distortion and distortion of timbre. As the sampling rate increases, listeners can hear more accurate details and more authentic timbre. In complex scenes such as video dubbing, games and singing that require a more refined and immersive sound experience, higher fidelity outputs, such as 48kHz sampling rates, will bring users an unprecedented new sensory experience.

Now, as the Azure deep neural network voice synthesis service upgrades the platform-wide AI sound to the 48kHz sampling rate, Microsoft is the first in the industry to bring a truly high-fidelity sound experience to AI voice users.

For more information on Microsoft Azure intelligent speech synthesis technology, click here.