Voice input for an APEX assistant fits neatly if the region lets you edit the transcript before it reaches the agent. Does the recognition happen in the browser or do you pass audio to a service, because that choice decides whether it works on the locked-down instances most of us deploy to. And when someone dictates something like "create a task for the Miller account", how much of that arrives as structured intent versus free text the agent has to guess at. What did you settle on?
In fact, the plug-in can use either the Web Speech API or a local AI model (https://huggingface.co/onnx-community/whisper-base) running in the browser using Transformers.js. In the latter case, no data leaves the browser: the transcription is performed locally, and the result is displayed in the AI assistant’s text box. The user can then review and correct it before clicking “Send.”
Adam Lewis
Product Engineer/Architect navigating the AI revolution
Voice input for an APEX assistant fits neatly if the region lets you edit the transcript before it reaches the agent. Does the recognition happen in the browser or do you pass audio to a service, because that choice decides whether it works on the locked-down instances most of us deploy to. And when someone dictates something like "create a task for the Miller account", how much of that arrives as structured intent versus free text the agent has to guess at. What did you settle on?