Skip to main content
Capture Signal AI and rule-based Signal insights from calls in English, Spanish, and French by tagging each call’s language and building language-aware Signals.
If you field calls in languages other than English, you can still capture Signal insights from the words spoken — using either rule-based Signals that spot specific phrases, or Signal AI to analyze calls for advanced insights. To do this, set up your Invoca account to capture which language a call is in — either automatically from the landing page that drove the call, or manually indicated by the caller. Once set up, Invoca can combine the same insight — such as “sales” — from each language into one Signal, tracking conversions together no matter which language they occur in.

Which languages and dialects are supported?

Currently, you can capture Signal AI insights in:
  • English (US)
  • English (UK)
  • Spanish (Spain)
  • French (France)

Step 1: Request Signal multi-language support

Reach out to your account team or support@invoca.com with the following:
  • Specify that you’d like to enable Signal multi-language support.
  • Indicate which of the above languages you’d like support for.
  • Choose a default language, used to process any call where the spoken language can’t be determined (see Step 2).
Once approved and processed, Invoca creates a new Marketing Data field in your account named “invoca_caller_language,” used to store the language value from your landing page or static marketing data assignments. You’ll use this field again in Step 2.

Step 2: Capture the language your calls are conducted in

Next, set up your account to capture the language spoken in each call. There are two options — you only need to follow one to capture AI Signals in your preferred languages, whichever fits your business best.

Option 1: Capture the caller’s language using an IVR keypress

  1. Log in to your Invoca account, click Campaigns, then select Manage Campaigns.
  2. Select the campaign you’re using. In the Call Treatment tab, click Edit.
  3. Follow the guide “Call Treatments & IVR” to create a new node in your IVR tree with the type “Ask a question.” Assign each keypress to a different language, and forward calls from each keypress to the destination number for that language.
  4. Note the numbers in brackets next to each keypress, and which number corresponds to each language — you’ll need this again shortly.
  5. Return to your Invoca network dashboard, click Settings, then in the Network section, select Marketing Data.
  6. Select the Tables tab, then click + Create New Table. Name it “Signal AI Language Reference.”
  7. Open the Lookup Field dropdown. If the language menu from step 3 is the only keypress menu in your IVR tree, select Key 1. If you have more than one, note whether this menu was the first, second, third, or fourth keypress menu top to bottom, and choose the matching Key number. Click Save.
  8. Click Download Template to get a template .xlsx spreadsheet. Open it and fill it out:
    1. Clear all data from column C onward.
    2. In cell B14, enter “Invoca Caller Language”; in cell B18, enter “invoca_caller_language.”
    3. In column B starting at row 19, enter the languages to capture, in the order listed in your menu — use “en-US” for US English, “en-GB” for UK English, “es-ES” for Spanish, and “fr-FR” for French.
    4. In column A starting at row 19, enter the bracketed numbers noted in step 4, matching each keypress.
  9. Save your spreadsheet, then upload it in the Upload Data window and click Upload.
  10. In the Tables menu of your Marketing Data dictionary, check the Last Run Summary column for your new table. A successful import shows “1 imported successfully.” If there’s an error, click the upload button in the Actions column to try again, or reach out to your account team or support@invoca.com for help.
  11. Repeat these steps for each additional language menu in your IVR tree, if any.

Option 2: Capture the language of your landing page with your Invoca Tag

To use this method, your landing page must store its language in a URL parameter, cookie, or JavaScript data layer — for example, a URL parameter like setLanguage=fr denoting the page (and any call driven from it) is in French. Once your landing pages store this information, follow the guide to capturing Marketing Data in your Invoca Tag, using these settings:
  • In the Attribution tile, find your “invoca_caller_language” field and enable it with the slider in the Enabled column.
  • In the Data Source Type column, select how your landing page stores its language — JavaScript, a URL parameter, or a cookie.
  • In the Data Source Name column, enter the exact name of the JavaScript layer, URL parameter, or cookie you use.
Click Save when finished.

Step 3: Create new Signals

Now that your account captures the language of each call, apply that data to your Signals. See the guides for creating rule-based Signals using Signal Builder and capturing advanced insights with Signal AI, and keep these tips in mind:
  • For rule-based Signals, phrase matching works against the transcript as spoken — not a translation to English. To capture Signals for appointments made on Thursday in Spanish, use “jueves,” not “Thursday.”
  • You can combine several phrase-spotting rules into one Signal to combine phrases from different languages — for example, “Thursday” OR “jueves” OR “jeudi.”
  • When creating a new AI Signal, you’ll see a Language dropdown near the Model Accuracy tile. Use it to select each language of call you’d like to analyze, and include training data from each of those languages when training your Signal. Invoca automatically combines insights for each language into the same Signal.

Step 4: Provide training data (Signal AI only)

Now that your AI Signals are created, train Signal AI to recognize when they occur. Currently, you can only train Signal AI in multiple languages using calls already recorded by your account — see the full guide on improving Signal AI accuracy. For your first version of a Signal, you’ll need a sample of at least 150 calls where the Signal did occur and at least 150 where it didn’t, for each language you want to analyze. You won’t need to sort calls by language yourself — Invoca identifies each call’s language automatically and sorts it into the correct training sample. Your account is now set up to capture language from your landing pages and analyze conversations for advanced, real-time insights in multiple languages. Check your Signal’s performance by clicking Signal in the sidebar menu, selecting Manage Signals, then opening your new AI Signal — you’ll find a dropdown to view data by language next to your performance meter.

FAQs

What if my landing page toggles between more than one language, or my customer visited more than one landing page in different languages? Invoca captures the last language your visitor selected and analyzes Signals based on that language. What if there’s an error and no language is captured? If your Invoca Tag doesn’t capture any language data, it assumes the call is in your default language and attempts to analyze it accordingly.

Where to go next

Last modified on September 23, 2026