Which image formats can Lyro extract text from?

Astica Read Text accepts HTTPS image URLs and Base64-encoded images, and works on printed text, handwritten notes, and images of varying quality. SVG and WebP are not supported, so a customer sending one of those formats will need to attach a different file.

Does Lyro get the position of the text it reads, or just the words?

Both. Astica Read Text returns detected text blocks with bounding box coordinates, so Lyro can tell which part of an image a value came from - useful on a receipt or a form where several numbers sit close together.

Can Lyro describe what is in a picture, or generate speech?

Not through this connection. Astica's wider platform covers computer vision and voice synthesis, but the connected toolkit exposes two actions - Astica Read Text for OCR and Analyze Audio for transcription - so Lyro extracts text and does not caption images or produce audio.

How does a customer's file actually reach Astica?

As a URL or a Base64 payload. Analyze Audio takes an audio URL or Base64 string and Astica Read Text takes an HTTPS image URL or Base64 image, so an attachment has to be reachable in one of those two forms before either action can run on it.

Astica.ai MCP integration

Pulls printed and handwritten text out of images a customer attaches, and transcribes audio clips into text the conversation can use.

Connect Astica.ai See all integrations

2actions available

What your agent can do with Astica.ai

Every action runs live through MCP. Nothing to build, nothing to maintain.

Astica_read_text
Lyro extracts text from images or screenshots a customer sends, turning visual information into searchable data. This handles product photos, invoices, or error screens instantly.
Analyze audio
Lyro processes audio files to transcribe and understand spoken customer input, unlocking voice channel interactions. This converts voice inquiries into text Lyro can respond to.

See all 2 actions

How businesses use Astica.ai + Lyro

Each card is one request a support team gets, and the Astica.ai actions Lyro runs to close it.

Read the text off a screenshot, receipt, or handwritten note
Lyro runs OCR on the image a customer attached and gets back the detected text blocks with their coordinates, so an order number or error message buried in a screenshot becomes a value the conversation can use.
Astica Read TextAnalyze Audio
Turn a voice note into text before anyone listens to it
Given an audio URL or a Base64 clip, Lyro transcribes what was said, so a recorded complaint or callback request is searchable and quotable instead of a file someone has to play through.
Analyze AudioAstica Read Text
Escalate with the attachment already transcribed
When a conversation needs a person, Lyro passes it over with the screenshot text and the audio transcript already in the thread, so the agent reads the evidence instead of opening and replaying each file.
Astica Read TextAnalyze Audio

How it works

Get started in 3 steps

Connect once, then just ask. There is no workflow builder to learn and nothing to maintain — Lyro reads the Astica.ai actions it has and picks the ones a request needs.

01
Connect Astica.ai
Authorize the Astica.ai account your team already uses — one consent screen, no API keys, no mapping tables. Lyro can only do what you granted that account, and you can disconnect it at any time.
02
Tell your agent what you need
Describe the job the way you would hand it to a teammate. Lyro maps it to the Astica.ai actions that close it and chains as many as the request needs.
03
Watch it work
The agent runs the actions inside the conversation the customer is already in, so nobody copies data between tabs and your team can take over at any point.
Get started free

Astica.ai · Lyro

Everything else about Astica.ai

Setup, permissions, and the limits of what Lyro can do inside Astica.ai.

Astica Read Text accepts HTTPS image URLs and Base64-encoded images, and works on printed text, handwritten notes, and images of varying quality. SVG and WebP are not supported, so a customer sending one of those formats will need to attach a different file.