Skip to content

A.I. text and drawing recognition

Do you already have the A.I. features add-on?

If you don't have it yet, you can purchase the A.I. features add-on on this website:

bsmagic.app/order/?buy=ai_features

Using A.I. recognition (OpenAI or Gemini)

In MysterSmith you can use A.I. based text and drawing recognition with GPT by OpenAI, Gemini by Google or Claude by Anthropic. To use these A.I. based recognition engines, you don't need to buy any text recognition credits from me, but you'll need your own OpenAI, Gemini or Claude API key — or a MysterSmith subscription (see below). I recommend using Gemini, as most of their models are FREE to use (except the Pro models).

The A.I. engines can recognize much more than text: in TEXT RECOGNITION → Recognition method choose text/numbers, playing card, celebrity name, city, animal, time, date, ESP sign, drawing keywords, analyze in detail, photo subject or your own custom prompt. All the settings: Text recognition.

FREE Gemini calls for subscribers

MysterSmith subscribers don't need their own API key: they get a number of free Gemini A.I. calls every day (and PREMIUM subscribers free Gemini image generations too). Just select Google Gemini (A.I.) as the recognition engine. If you also enter your own Gemini key, your key is used first and the free calls are the backup — unless you switch on Use FREE Gemini calls first in the A.I. FEATURES section. More: Free Gemini A.I.

Obtaining your OpenAI API key

To be able to use GPT-based A.I. text recognition (also drawing recognition and analysis), you will need to get an OpenAI API key, and top up your balance (performing one routine will cost you only around 0.01-0.02 USD).

Follow these steps to get your OpenAI API key, to use in the MysterSmith app:

IMPORTANT: copy your API key as soon as you create it, you won't be able to view/copy it later!!!

  • make sure you have a username and PIN set up in MysterSmith (Account Settings - Show details)
  • in MysterSmith go to Settings, and scroll down to the A.I. FEATURES section
  • paste your API key to "OpenAI API key"
  • select the model at "OpenAI engine" (in the A.I. FEATURES section)
  • in TEXT RECOGNITION settings select "OpenAI GPT (A.I.)" as the recognition engine
  • set "Automatic recognition" to 3 sec, and enable "Show recognized text (pop-up)"
  • tap START ROUTINE and start the Doodle screen
  • write anything on the screen (or use an impression board)
  • the recognized text will pop up at the bottom of the screen
  • you can also try automatically performing an action after text recognition (see the screenshots and photos below)

How to create your Gemini API key

You can also use A.I. text and drawing recognition provided by Google Gemini, which can be used FREE of charge (up to a specified limit, except the Pro models).

Here are the steps to create your Gemini API key, and to use it in MysterSmith:

  • visit the Google A.I. Studio: https://aistudio.google.com/apikey
  • accept the Terms of Service
  • select API keys - "Create API key"
  • choose an existing project, or select "Create API key in new project"

IMPORTANT: copy your API key as soon as you create it, you won't be able to view/copy it later!!!

  • make sure you have a username and PIN set up in MysterSmith (Account Settings - Show details)
  • in MysterSmith go to Settings, and scroll down to the A.I. FEATURES section
  • enter your Gemini API key
  • select the model at "Gemini A.I. engine" in the A.I. FEATURES section (a Flash-Lite model is a good start)
  • in TEXT RECOGNITION settings select "Google Gemini (A.I.)" as the recognition engine
  • set "Automatic recognition" to 3 sec, and enable "Show recognized text (pop-up)"
  • tap START ROUTINE and start the Doodle screen
  • write anything on the screen (or use an impression board)
  • the recognized text will pop up at the bottom of the screen
  • you can also try automatically performing an action after text recognition (see the screenshots and photos below)

In Google A.I. Studio you can view the usage (and optional billing) reports. Using Gemini A.I. recognition with your own key is FREE up to a limit set by Google (a number of recognitions per minute and per day — see the current limits in Google A.I. Studio; the Pro models are paid).

Using Claude

You can also use Claude by Anthropic for recognition (and for A.I. text prompts and acronyms):

  • create an account and top up your balance in the Claude Console: https://console.anthropic.com
  • create an API key (Settings → API keys), and copy it right away
  • in MysterSmith go to Settings → A.I. FEATURES, paste it to "Claude API key" and select the model at "Claude A.I. engine"
  • in TEXT RECOGNITION settings select "Claude (A.I.)" as the recognition engine
  • set "Automatic recognition" to 3 sec, and enable "Show recognized text (pop-up)"