der3318 / audio-powered-gpt Goto Github PK

A tiny WPF interface that integrates Azure cognitive service with GPT endpoint. This requires Azure subscription resources of both speech service and OpenAI.

Simply type or speak (via microphone) to ask GTP questions in this mode. Press the "start button" to trigger a speech QA session, and click the "start/stop button" again to pause.

This is the real time translation (into Chinese) functionality. Result texts will also be displayed as a 3-second toast in the bottom corner, so the app can be run completely in the background.

An audio redirection (from speacker to input) interface is a prerequisite to use the feature. Windows stereo mix or VoiceMeeter is probably a good choice.

• Icon: https://arstechnica.com/information-technology/2023/01/openai-and-microsoft-reaffirm-shared-quest-for-powerful-ai-with-new-investment/
• Azure Speech to Text: https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/how-to-recognize-speech
• Azure OpenAI Studio: https://learn.microsoft.com/en-us/azure/cognitive-services/openai/quickstart
• Toast Notification: https://learn.microsoft.com/en-us/windows/apps/design/shell/tiles-and-notifications/send-local-toast
• Embedded WPF Markdown Viewer: https://github.com/whistyun/MdXaml