Microsoft announced the public preview release of Azure AI Speech.
This digital product, developed by the mentioned company, is a technology with which consumers can create videos with talking avatars. To solve the corresponding task, users must enter text and create an interactive bot in real time, applying images of people.
The avatar of text-to-speech transformation is a new feature with visualization capabilities, that allows consumers to generate synthetic videos with a talking 2D photorealistic character. This information is contained in a message that was published on the Microsoft blog last Wednesday, November 15.
The process of training the avatar model is carried out using deep neural networks based on samples of videos in which a person figures. The voice in this case is provided by a special system for converting text material into speech.
A Microsoft blog post claims that the avatar, which is capable of generating speech based on text, can be used for training videos, advertising, product presentations, virtual human resources assistants, digital sales agents, artificial intelligence teachers, and other apps and applying scenarios.
Avatar videos usually consist of elements typical of such content, including character activity, which in this case is a talking digital image, background picture, and ambient music.
The company stated that one of the main reasons for building a new product is to simplify the creation of video materials. In this case, traditional methods involve significant costs in terms of time and financial investments in shooting and editing. With the avatar offered by the tech giant, consumers can enter text to generate videos according to their needs.
The release of the Azure OpenAI service and the neural transformation of text materials into speech have improved what can be described as the level of naturalness of interactive conversations. The text transformation Avatar provides consumers with the opportunity to generate fascinating digital interactions. This product based on artificial intelligence is great for creating virtual assistants, dialog agents, chatbots, and much more.
Azure AI Speech offers two functions for converting text materials into speech. A ready-made avatar provides ready-made products. In this case, consumers have the opportunity to choose from a variety of video content options and interactive apps. The text-to-speech feature allows users to create personalized avatars for products and brands by uploading their videos.
Microsoft is committed to the responsible use of artificial intelligence. The tech giant provides access to avatars only in case of registration for certain applying scenarios. This approach ensures the protection of individual and public rights. Also, in this case, the prevention of malicious deepfakes is guaranteed and blocks the possibility of creating information materials that may mislead.
The interactive avatar uses the Azure Open AI Service GPT-3.5 model to process consumer requests. The technology also gets access to outdoor shop data sources to answer questions about the product range or customer accounts.
As we have reported earlier, Microsoft Launches AI-Powered Service for Blind Customers.