Add Cloud-Based TTS Voices as System-Wide Windows Voices

[harshid86]

Description of the new feature / enhancement

The feature would allow users to register cloud-based Text-to-Speech (TTS) services, such as Azure Cognitive Services or OpenAI's TTS API, as system-wide Windows voices. These voices would appear in the Windows Speech settings and would be selectable in any application that supports the Windows Speech API, alongside built-in voices like “Microsoft Hazel.”

Users would configure the feature by providing API credentials and selecting their preferred voice from the cloud service. The TTS audio would be fetched in real-time from the cloud and processed for use by any TTS-enabled application.

Scenario when this would be used?

This feature would be useful in scenarios where:

1. Users need higher-quality, natural-sounding voices for accessibility tools like screen readers.
2. Multilingual users require advanced language support that built-in Windows voices might lack.
3. Power users want to customize the TTS experience with advanced AI voices for productivity, such as voice-assisted workflows.

For example, a visually impaired user could benefit from hearing documents read aloud in a more natural and expressive voice, making the listening experience less monotonous. Similarly, professionals working in different languages could use advanced AI-generated voices for better pronunciation and clarity.

Supporting information

This idea builds on PowerToys' existing integration with OpenAI for its advanced paste functionality, demonstrating that cloud-based services can be integrated effectively into the toolset.

[harshid86]

Additional considerations:

• The feature would require an internet connection and user-provided API keys for the chosen service. If the cloud service is unavailable, default to built-in Windows voices to ensure continuity.
• Allow users to set a maximum monthly budget for API usage.