I wonder how it would go if you built a little interpolation agent to take in what you say raw, reformat it with good grammar, and then print that reformed dictation to whatever app you were using.
Right now, I've actually set this up with a keyboard shortcut! I had Claude walk me through how to build it with Automator on Mac, and using ollama to run a llama instance locally on my laptop. So when I copy the text that I've dictated and hit my keyboard shortcut, it takes what's in my clipboard and replaces it with a version of text interpreted by the LLM.
I wonder how it would go if you built a little interpolation agent to take in what you say raw, reformat it with good grammar, and then print that reformed dictation to whatever app you were using.
Right now, I've actually set this up with a keyboard shortcut! I had Claude walk me through how to build it with Automator on Mac, and using ollama to run a llama instance locally on my laptop. So when I copy the text that I've dictated and hit my keyboard shortcut, it takes what's in my clipboard and replaces it with a version of text interpreted by the LLM.