Microsoft has recently added a new Copilot Audio mode to its Copilot tool, which is based on its self-developed MAI-Voice-1 model. Users can now choose from three different voice modes to meet the needs of different scenarios.
The first is the "Emotional Mode," which allows expressive free-form expression and is suitable for situations that require emotional delivery; the second is the "Story Mode," which allows users to tell stories with multiple characters, offering an engaging auditory experience; and the third is the "Script Mode," which provides word-for-word precise playback, making it ideal for situations where accurate information delivery is required.
In the new audio mode, Microsoft offers a variety of voices and styles, ranging from classic Shakespearean readings to passionate sports commentary. This diversity not only provides users with a wide range of choices but also greatly enhances the fun and personalization of the interaction.
Currently, this new feature is available in Copilot Labs, allowing users to directly experience this innovative voice interaction method. Notably, Microsoft's recently launched MAI-1 is its first large language model, and it has also reached an agreement with Anthropic to integrate its model into Office applications. These initiatives indicate that Microsoft is striving for greater independence from OpenAI, enabling more flexible development and application of artificial intelligence technology.
Experience entrance: https://copilot.microsoft.com/labs/audio-expression
Key points:
🌟 Users can choose between emotional, story, and script audio modes in Copilot.
🎭 The new feature offers various voices and styles, suitable for different expression needs.
🔍 Microsoft's MAI-1 model marks its independent development in the AI field.
