At its annual Connect conference, Meta announced that it has expanded the capabilities of its AI assistant, Meta AI, by adding voice and vision support. With this move, Meta AI will now be able to understand spoken commands as well as describe images. This enhanced version directly competes with the Advanced Voice Mode of ChatGPT, which was just introduced recently by OpenAI.
Table of Contents
Key Features of Meta AI with Voice Mode
1. Voice Interactions
Meta AI can now understand and respond to voice queries instead of only text inputs. Users can speak naturally without following prompts.
2. Celebrity Voices
The voice of Meta AI can be selected from celebrity options like John Cena, Dame Judy Dench, Kristen Bell etc.
3. Vision Support
In addition to voice, Meta AI has also gained the ability to describe images shared by users in conversations. Plus, Meta AI understands images shared in chats and can answer visual questions and edit photos by commands.
4. Platform Integration
The voice feature will be rolled out on Messenger, Instagram, WhatsApp and Facebook starting this month in key markets.
5. Scale and Reach
Meta aims to make Meta AI the most used AI assistant globally with its integration in main apps and services. The company claims nearly 500 million monthly users.
New Llama Models by Meta AI
1. Llama 3.2 Models
Meta’s AI research models have been upgraded to version 3.2 with multi-modal vision and language capabilities. The new 11B and 90B Llama models can understand images, charts and captions.
2. Lightweight Models
Meta also introduced Llama 1B and 3B for use in privacy-conscious European markets on smartphones and edge devices.
Meta AI Voice vs Advanced Voice Mode of ChatGPT
OpenAI’s ChatGPT with voice launched to appreciation but had limited access. Meta aims to provide its AI services freely to a much wider community of users globally and across product categories. Meta’s move directly takes on OpenAI’s popular Advanced Voice Mode feature for ChatGPT. Both systems now offer more natural speech-based interactions.
| Related: OpenAI Introduces Advanced Voice Mode for ChatGPT With 5 Unique Voices
Demo and Response Quality
Meta CEO Mark Zuckerberg demonstrated the functionality on stage, asking Meta AI questions using Awkwafina’s voice. While responses were satisfactory, the cadence was less conversational than ChatGPT, as per the report. Voice remains a more intuitive way to interact with AI than text, giving Meta AI an edge over text-only systems to gain a user base.
Head-to-Head with OpenAI
By introducing voice capabilities similar to ChatGPT, Meta provides an alternative for users. Both systems offer a more natural way to interact with AI over traditional text assistants. It is still unknown which voice-powered virtual assistant dominates the market.
| Latest From Us
- Hugging Face CEO Shares His 2025 AI Predictions
- Stanford 2024 AI Index Report Confirms That AI Leaves Human Capabilities in the Dust Across Domains
- World Labs Introduces Spatial AI Model That Lets You Navigate 3D Worlds from 2D Images
- Nous Research Develops DisTrO Powered by Distributed Machines Across the Internet
- Tencent Introduces HunyuanVideo, An Open-Source Triumph in Video Generation Excellence