Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Creating an AI Helper to add speech input to any windows app!
Demonstrates using low‑level Windows APIs to add local or cloud speech‑to‑text and LLM integration to any existing Windows application without source changes.
I will show a novel way to use low level Windows API to integrate local (or cloud-based) speech-to-text model for adding speech input to applications that don’t natively support it on Windows.
Existing applications we use today were built before modern AI and LLMs were available. Adding Voice input to an existing application without
making any changes to that application is an enabler for all of us.
You can use it entirely locally on your PC if you like, so it is private and secure. I will also demo how it can be used with Multi-Modal LLMs to assist you with more advanced tasks. This can be extended to use local and/or open source language models like Qwen and GPT-OSS. This is open source and available for anyone to use, tinker with and apply to other unique problems.
SPCHR transcribes real-time speech using Azure Speech or local Whisper.net.
SPCHR: Windows speech-to-text application leveraging Azure Speech or local Whisper.
Compose Email
Loading recent emails...