[Suggestion] Semantic Kernel voice transcription plugin (.NET/C#)

[deepgram-robot]

Integration: Microsoft Semantic Kernel (.NET)

<!-- metadata
type: queue
slug: 470-semantic-kernel-voice-plugin-dotnet
language: csharp
products: stt|tts
priority: user-request
-->

What this should show

A Deepgram plugin for Microsoft Semantic Kernel that exposes STT (pre-recorded transcription) and TTS (Aura) as [KernelFunction] attributes. Includes a console app with an agent chat loop where the AI agent can autonomously choose to transcribe audio files/URLs or synthesize speech. Demonstrates how .NET AI agent developers can integrate Deepgram voice capabilities into their Semantic Kernel pipelines.

Credentials likely needed

• DEEPGRAM_API_KEY
• OPENAI_API_KEY (or Azure OpenAI key for the Semantic Kernel LLM backend)

---

Original request:

What to build

A working example demonstrating Deepgram as a Microsoft Semantic Kernel plugin for .NET AI agents — enabling voice transcription as a kernel function that AI agents can invoke to process audio inputs.

Why this matters

Microsoft Semantic Kernel (53k+ stars) is the dominant .NET AI agent framework, used by enterprise teams building AI assistants and copilots. Developers building voice-enabled .NET agents need a reference integration showing Deepgram as a Semantic Kernel [KernelFunction]. There is currently no example of Deepgram + Semantic Kernel working together, leaving .NET AI agent developers without a clear path to add speech capabilities.

Suggested scope

• Language: C# / .NET 8
• Framework: Microsoft.SemanticKernel (latest stable)
• Deepgram APIs: Pre-recorded STT, TTS (Aura)
• What it does: Semantic Kernel plugin class with [KernelFunction] attributes for TranscribeAudio (file/URL → text) and SpeakText (text → audio). Includes a sample agent chat loop that can process voice commands.
• Complexity: Medium — plugin class + sample console app

Acceptance criteria

• Runnable with minimal setup (clone, add API key, run)
• README explains the pattern clearly
• Uses current Deepgram .NET SDK and Semantic Kernel versions
• Plugin functions are properly annotated with descriptions for AI agent discovery
• Includes example of agent autonomously choosing to transcribe audio

---

Raised by the DX intelligence system.

[github-actions]

Ooo, we'll get right on that! 🎉

The Engineer will pick this up shortly and build a Semantic Kernel voice transcription plugin (.NET/C#) example — showing Deepgram STT and TTS as [KernelFunction] attributes that an AI agent can invoke autonomously.

I'll keep this issue open so you can track progress — we'll close it automatically when the PR merges.

If you have any extra context (preferred language, specific API, or credentials you're already using), drop it here — it helps the build go faster!

[github-actions]

Built in #191