Greetings,
I’d like to request a new feature as a paid user: multimodality well embed into Cursor.
On a first trial, development speed seems boosted by Cursor current capabilities, but there’s margin for further improvement if well integrated with voice input/command capabilities. This is somewhat related to the following past threads (on a quick search):
These threads (and a few more, in fact) offer/demand such capabilities, but a proper and well-engineered voice commanding integration is not available as far as I know. This would imply commanding text generation across code and Chat in a good-enough human-machine interaction flow. Having this ready soon would be a good reason to consider further upgrading our memberships. Ping @truell20, @amanrs (and other admins).
Edit: It would be fantastic if the multimodality could also interact with the in-Cursor Terminal.
4 Likes
Greetings, I’d like to insist on this regard. This feature is very much needed and would further justify extending the paid accounts in our group to more developers, as well as securing the existing ones.
Please provide an intuition as well as a timeline of the feasibility and expectations.
1 Like
I’d like to add that we will hopefully soon (“soon” knowing them) have access to OpenAI’s API for their low latency and expressive advanced voice mode with GPT4o. I would love the option to be able to use my API key to integrate this with cursor. If extra help is needed implementing this please reach out once the API is available; I have been dreaming of a feature like this for a long time so I will probably be trying to implement it myself anyway if it doesn’t come fast enough.
1 Like
Agree. Hopefully in a few years typing will be redundant, you could go on a trip in your AR glasses, talk to the ai and it will code for you. I suppose it would be beneficial for cursor to start jumping on this trend early on, so that once those capabilities are out, users would flock to it to reap the benefits of keyboardless workflows.
1 Like