Skip to content
AskVault Docs
Try Free →

Photos and voice notes on WhatsApp

Rolling out. This is not switched on yet. Until it is on for your workspace, photos and voice messages work as described in Voice messages, photos and files.

A customer can photograph the part they need, or its label, and send a short voice note in their own language. AskVault reads the part number printed on it and looks the product up. When it finds the part in your Shopify or WooCommerce store, it checks live stock. It replies in the language the customer spoke, with a Buy now button when the part is in stock, or a View product button otherwise.

What the customer gets

  1. They send a photo. AskVault reads any part number, model number or SKU printed on the item, its label or its packaging. It is told to skip any code it can't read clearly rather than guess one, so a clear, close photo of the label works best.
  2. AskVault finds the part. It looks for an exact match on the SKU in this order:
    • your Shopify store, with stock checked live,
    • your WooCommerce store, with stock checked live,
    • the products AskVault found when it read your website. Stock there comes from your product pages, so the reply says it is listed on the website and does not promise it is in stock.
  3. They get an answer and a button.
    • In stock, from Shopify or WooCommerce: a Buy now button that opens checkout with the part already in the cart.
    • Out of stock, stock unknown, or found on your website only: a View product button, when the product has a web page.
    • If your WhatsApp number has a Meta product catalog, the customer gets the native WhatsApp product card instead, and can add it to their WhatsApp cart.
    • No match: the reply says so plainly and asks for the part number on the label or a sharper photo. No button is sent.
  4. They send a voice note. AskVault writes down what they said, in the language they spoke, and answers in that language. A voice note sent within 15 minutes of a photo can refer to it, for example "I need two of these", and the button opens checkout for two.

Each photo and each voice note gets its own reply. Button labels are translated for English, Hindi, Bengali, Tamil, Telugu, Marathi, Gujarati, Kannada, Malayalam, Punjabi, Urdu, Spanish, French, German, Portuguese and Dutch; other languages see them in English. If your workspace's reply language is set to always use one language, replies and buttons use that language.

What your team sees

In the inbox, a photo shows as [image] followed by any caption the customer typed. A voice note shows as [voice message] followed by what AskVault heard, so a mis-heard word is never mistaken for something the customer typed. The part numbers AskVault read, and the language the customer spoke, are saved with the conversation's customer details. The photo and the recording themselves are not stored.

Requirements

  • WhatsApp connected through Meta (the WhatsApp Business Platform). Numbers connected through Twilio keep the behaviour in Voice messages, photos and files.
  • A plan that includes WhatsApp (currently Growth or higher).
  • For a Buy now button: Shopify or WooCommerce connected under Integrations. Stores AskVault only read from your website get a View product button.

Privacy

A photo is sent to AskVault's AI provider to be read. A voice note goes to the same AI provider, and to our speech-to-text provider if the AI provider can't transcribe it. AskVault doesn't store either one.

Message credits

Each answer uses message credits the same way a typed question does. Reading the photo or voice note doesn't cost extra credits.

Was this page helpful?