New Delhi,UPDATED: Nov 17, 2023 3:06 PM IST
Microsoft has introduced a new tool, the Azure AI Speech text-to-speech avatar. This allows users to create realistic avatars that deliver scripted content. Unveiled at the Microsoft Ignite 2023 event, this new feature is currently in public preview and offers a unique way to generate videos with photorealistic avatars that say text-based content.
Users can upload images of their desired avatar appearance and provide a script for the virtual character to speak. Microsoft’s tool uses a model to animate the avatar, while a separate text-to-speech model, pre-built or trained on the person’s voice, speaks the script. This technology allows users to efficiently create videos for various purposes, such as training sessions, product introductions and customer testimonials, using simple text input.
The avatars generated by this tool can communicate in multiple languages, and in chatbot scenarios it can use AI models such as OpenAI’s GPT-3.5 to respond to unscripted customer queries. Additionally, Microsoft says it is cautious about the potential misuse of such technology. To address this, most Azure subscribers only have access to pre-built avatars, while custom avatars are limited to specific use cases, which require registration and approval.
This precaution is in line with Microsoft’s commitment to protect individual and societal rights, encourage transparent human-computer interaction, and combat the spread of harmful deepfakes and misleading content. Microsoft emphasizes that the text-to-speech avatar is designed with responsible use in mind.
To further ensure responsible use, Microsoft has established specific requirements for users of custom avatars, TechCrunch reported. Customers must obtain express written consent and consent notices from avatartalent describing duration, use, and any content restrictions. Additionally, customers are required to include disclosures stating that the avatars are created with AI and generated by AI.
The introduction of this tool not only adds an exciting dimension to digital content creation, but also underlines Microsoft’s approach to ethical considerations. As concerns about deepfake technology escalate, Microsoft’s proactive approach aims to strike a balance between technological innovation and responsible use, and set a standard for AI-powered content creation tools.