How To: Talk to GPT-3 through Siri

Note: it’s now possible to talk to a newer OpenAI model (gpt-3.5-turbo) through Siri – if you want to use the newer, and much cheaper, version, see my updated post here. The new version seems to do better on some questions but worse on others.

Like many others, I’ve been incredibly impressed with OpenAI’s ChatGPT and how far language models have come since I was working on natural language processing research a few years ago.

But, also like many others, I’ve been regularly frustrated with Apple’s Siri and how it often fails to give useful answers to even the most basic of questions. This week, after a few too many unsatisfying Siri answers in a row, I started to wonder if it would be possible to solve the problem once and for all by querying ChatGPT directly through Siri.

It turned out that while the ChatGPT API isn’t officially available yet to pose questions to programmatically, the recent GPT-3 model is, and it’s also very powerful.

It also turned out that several people have written Siri shortcuts to interface with GPT-3, but I had a couple of issues with them when I tried them out:

  • Siri wouldn’t always read the answers out loud
  • The answers often started with a stray question mark, which was distracting when Siri did read the answer out loud, starting with “Question mark…”

I tried fixing the first issue by making a shortcut that included steps to explicitly read the answer out loud, but it turned out that a far simpler solution was just to change Siri’s settings in Settings->Accessibility->Siri->Spoken Responses to “Prefer Spoken Responses”:

Make sure you turn on “Prefer Spoken Responses” if you want to use this shortcut and have Siri read the answers out loud to you!

The stray leading question marks were an easier fix – I just modified an existing shortcut with a step that strips them out.

With those two fixes, the shortcut started working very seamlessly – I can now tell my phone “Hey Siri, GPT Mode”, then a question, and quickly get a response from GPT-3 read back to me by Siri.

You can download the Siri shortcut here to add it to your phone (and you can see the original shortcut that I modified it from here).

The shortcut itself is free to use; you’ll just need to create an OpenAI account here, create an OpenAI API key and paste it into the text field in the shortcut that says “Replace this with your OpenAI API key!” You can see more detailed instructions for how to do this in this article explaining how to use a similar shortcut. Right now, OpenAI is offering three months of API credits for free when you sign up for an account, and they don’t charge much once you run out of your free credits.

Once you’ve installed the shortcut onto your phone and pasted your OpenAI API key into it, it’s ready to use! You can use it under the name I gave it (“GPT Mode”) or rename it to your taste, to anything that doesn’t conflict with Siri’s existing commands.

To use the shortcut, say “Hey Siri”, then “GPT Mode” (or whatever you renamed the shortcut to), then say whatever you want to ask GPT-3. Siri will then, after a delay, display the answer and read it out loud.

Much like with ChatGPT, the answers aren’t always right, but they usually are, and I’ve found that the ability to ask my phone complex questions and get reasonable answers back is incredibly useful as long as I keep the imperfections of GPT-3 in mind.

Hopefully, the day is coming soon when Siri will have this kind of functionality built in natively! In the meantime, there’s this Siri shortcut.

PS – No pressure, but if you’ve found this shortcut useful, I’d appreciate it if you buy me a coffee!

55 thoughts on “How To: Talk to GPT-3 through Siri

  1. This is cool, but I think I have it misconfigured, because siri will take precedence over the shortcut a lot of the time.
    Any time I ask it a question, for example, Hey Siri, GPT Mode, How old is alec baldwin? I get siri’s main question answering interface, not a ChatGPT text response.

    Since almost everything I ask is a question of the form “What is…?”, “How is…?”, or “Why is…?”, almost no questions I ask go to ChatGPT.

    Like

  2. This is cool, but is pretty unusable for one main reason. Any time I ask a question of a certain form, Siri tries to take over and answer it with her own software, and won’t pass it to the OpenAI endpoint.

    For example, if I say “Hey Siri, GPT Mode, How old is Alec Baldwin”, Siri automatically takes over, and answers, and the shortcut does not proceed. This happens with any question that starts with How, Why, or What, which is obviously the majority of questions!

    Is my Siri misconfigured or does this always happen with this shortcut?

    Also, I notice that when I manually run the shortcut, by clicking on it from the shortcuts app, I don’t run into this problem, the problem seems to have something to do with the fact that siri is still listening for questions at the same time that the dictation input command runs, and decides to take precedence based on the content of the question.

    Like

    1. Hey Nate, yeah, I’ve been having the same issue with Siri taking over for questions it thinks it knows how to answer. Phrasing the questions differently usually helps but if you find a way to stop the Siri takeover I’d appreciate you leaving another comment so everyone sees the fix. Thanks for the comment.

      Like

  3. This works very well. So far I make pauses between Siri – GPT Mode and wait until Siri asks” what’s the text?”

    Is there anyway to ask Siri to make a note of the response or message it to someone ?

    Like

    1. I haven’t run into that one myself, but it’s worth checking if you have credits on your OpenAI account and whether you’ve pasted the API key correctly into your shortcut (I accidentally deleted the hyphen when I was first pasting it over but caught my mistake before trying to actually use it.)

      Like

    2. I have the same problem. I get the initial enquiry window, then shortcut runs but there is no returned info window.
      I have API credits, tried a new key and approved all permissions for the shortcut.
      Any suggestions most welcome as this seems a neat solution

      Like

  4. Thanks for posting this shortcut. It worked for me. The first time I asked Siri to run GPT mode Siri asked permission to run the shortcut, then it asked permission for the API to access the internet. After I gave both permissions, I was able to ask again by pausing slightly following the trigger phrase “Hey Siri”, “GPT mode” then wait for the response “What’s the text?” and then say my prompt.

    There was a delay of a few seconds before my prompt “What are the benefits of the La Niña weather system?” was answered with a seven point bulleted response. Finally, a more useful Siri!

    Liked by 1 person

  5. Could you please help us with the steps needed to get rid of the annoying question mark that shows up before every response

    I’d really appreciate it!

    Like

      1. Hi. Thanks for the link. I can ask the question and it takes the input but I don’t receive any output? Am I missing something. I thought it might be the api but even if I create another one it does the same.

        Like

      2. Hi Elizabeth, it could be a permissions issue. Open shortcuts app on iPhone and tap on your GPT Mode shortcut. You may then see a dialogue asking for permissions, followed by another one asking for permission to access the internet. Say Ok both times then you should be able to receive responses from just prompting via Siri. See my earlier post for my experience.

        Like

  6. Hey there!
    I appreciate the shortcut but sometimes I prefer writing the prompt, is there any way to achieve that ? Thank you lots in advance

    Like

    1. To have the choice between wring or speaking the prompt download the new Bing app for iPhone. This uses GPT-4 and is also connected to the internet so answers are current. You may need to first get approval from Microsoft to access the “new Bing” before the app will work with the chat functionality. Approval for those with a Microsoft account is pretty quick.

      Like

  7. When asked date and time it gives weird reply. like 1988 Nov 16. same happens time. it gives USA time instead of my country which is 11 hours ahead of US time. Can’t we run GPT-4 in siri?

    Like

  8. Two questions I know others may be thinking:

    1. Do you HAVE to wait for “What’s the text?” prompt from Siri before asking the question?

    2. If you do have to have the prompt, can you edit it to some other form of question?

    Like

      1. Ah OK, but where in the shortcut do you edit that text? I don’t actually see it anywhere – just the “Dictate text” step – no reference to that phrase “What’s the text?”

        Like

Leave a comment