Resources · Make integration

AssemblyAI Make integrationAutomate AssemblyAI with Make.

Got recordings piling up that nobody has time to type out? The AssemblyAI Make integration gives you 16 modules: one instant trigger, 14 actions and one search. This guide is for teams who live in calls, podcasts or interviews, and it walks you from connection to a working first scenario.

Try the AssemblyAI integration on MakeSee the 16 modules

Verified Trustpilot reviews · AI, automation & growth agency

What you can automate

What does the AssemblyAI Make integration let you automate?

The AssemblyAI Make integration links your AssemblyAI account to Make, a no-code tool where you build a scenario: a chain of modules that runs on its own. Each module is one brick (a trigger, an action or a search), so an audio file (a call, a podcast, an interview) can travel from the app that stores it to a transcript, the written text of that recording, then on to wherever your team reads it.

Subtitles without the typing. A video producer drops a file, Transcribe an Audio File turns it into text, and Get Subtitles for a Transcript hands back an SRT or VTT file ready for the editor.

Call notes nobody has to write. A sales team records its demos. Run a Task using LeMUR sends the transcript to a language model with your own instruction, for example a summary or the list of objections raised, and Make posts the answer where the reps work.

A keyword alarm on interviews. A research team wants to know which recordings mention a competitor or a product name. Search for Words in a Transcript scans a completed transcript for the words you list, and the result can feed a spreadsheet or a message.

The official documentation publishes no limits for this app, so plan around Make itself: on the Free plan, files stop at 5 MB, which a long recording can exceed. If your team already works in n8n, the AssemblyAI n8n integration covers the same app, and n8n vs Make lays out how the two tools differ. New to the editor? The Make training starts from zero.

Connect

How do you connect your AssemblyAI account to Make?

  1. 01

    Add an AssemblyAI module

    In your scenario, click the +, type AssemblyAI in the search box and pick any of its modules. Then click Create a connection. A connection is your account linked to Make once, and every AssemblyAI module reuses it.

  2. 02

    Name the connection

    This step is optional, but a clear name pays off later. If a colleague and you both link an AssemblyAI account, a name like "AssemblyAI podcast team" tells you at a glance which one a module is using.

  3. 03

    Authorize and save

    Follow what the window asks: either authorize Make on the AssemblyAI page, or paste the key your AssemblyAI account gives you. Click Save. The module now shows your account, and its fields can list what Make is allowed to read.

First scenario

Your first scenario: turn a new recording into a transcript

GoalWhen a new audio file shows up in the app where your team stores recordings, Make sends it to AssemblyAI and brings the finished transcript back.

  1. 01

    Create the scenario

    From the Scenarios page, create a new scenario and click the +. Pick the trigger of the app that receives your recordings, such as a cloud storage folder or a form with a file field.

  2. 02

    Add Upload a File

    Click the + on the right of that first module, search AssemblyAI and choose Upload a File. Map the audio file from the first module. AssemblyAI keeps it private and removes it once the transcription is done.

  3. 03

    Add Transcribe an Audio File

    Add Transcribe an Audio File and map what Upload a File returned. The names suggest this link; check it with Run once. This module waits on its own until the transcript is completed or in error.

  4. 04

    Test with Run once

    Drop a short test recording in your source app, then click Run once. Each module shows the bundles it received (a bundle is one item passed along, here one file and then one transcript). Read the text before going further.

  5. 05

    Schedule and switch on

    If your first module is a scheduled trigger, open the clock on it: a new scenario checks every 15 minutes by default. Then switch the scenario on. Every module run counts as one operation on your Make plan.

Triggers

What can start a scenario from AssemblyAI?

1 module starts a scenario when something happens in AssemblyAI. A scheduled trigger is checked by Make on the scenario's schedule; an instant one is called by AssemblyAI as soon as it happens.

AssemblyAI1Watch for Transcript Ready Notification

Watch for Transcript Ready Notification

Instant trigger

In Make“The notification when the transcript status is completed or error.”

This trigger starts your scenario the moment AssemblyAI has finished with a transcript, whether the job ended as completed or as an error. You get one bundle per finished transcript, ready to pass to the next module.

How it fires
instant, AssemblyAI calls Make through a webhook as soon as it happens. A webhook is an address Make gives AssemblyAI so it can ring the bell itself: no clock to set.
When to use it
when transcripts are started somewhere else and you only want to react once the text exists, for example to route it to a notes app.
Watch out
error results arrive too, so add a filter on the status before sending anything to a colleague.
Modules

What can Make do in AssemblyAI?

AssemblyAI gives you 15 modules. For each one: what it does for you, when to reach for it, and what to watch out for.

AssemblyAI1

Delete a Transcript

Action

In Make“Deletes a transcript by ID. Deleting a transcript does not delete the transcript resource itself, but removes the data from the resource and marks it as deleted.”

Wipes the content of one transcript, found by its ID. The record stays in your account, flagged as deleted, but the text and data behind it are gone.

When to use it
at the end of a flow handling sensitive calls, once the useful output (a summary, a ticket) is saved elsewhere.
Watch out
there is no undo, so place this module after every step that still reads the transcript.
AssemblyAI2

Get a LeMUR Response

Action

In Make“Retrieves a LeMUR response that was previously generated.”

Fetches an answer LeMUR already produced, so you can reuse it without asking the language model again. You point it to the earlier request and get the stored output back.

When to use it
a second scenario that publishes or archives answers generated by a first one.
Watch out
once a LeMUR request has been purged, its answer is no longer there to fetch.
AssemblyAI3

Get a Transcript

Action

In Make“Retrieves a transcript by ID.”

Pulls one existing transcript by its ID, with everything AssemblyAI holds for it. It is the plain way to bring text back into a scenario that did not create it.

When to use it
after the instant trigger, to load the full transcript it announced before sending it to a document or a CRM.
Watch out
the ID must come from an earlier module; mapping it keeps the flow working for every file.
AssemblyAI4

Get Paragraphs of a Transcript

Action

In Make“Retrieves all paragraphs of a transcript.”

Returns the transcript cut into paragraphs instead of one long block of text. Readers find a paragraphed interview far easier to skim.

When to use it
when the text lands in a blog draft, a document or a knowledge base where layout matters.
Watch out
check with Run once whether paragraphs arrive as one bundle or several; if several, a Make aggregator joins them into one field.
AssemblyAI5

Get Redacted Audio of a Transcript

Action

In Make“Retrieves the redacted audio of the transcript.”

Retrieves the redacted version of the audio linked to a transcript, meaning the recording with the masked passages instead of the original file.

When to use it
when a recording must be shared outside the team, with a partner or a trainer, and the raw file should stay internal.
Watch out
run it once on a real transcript and listen to the result before automating the sharing step.
AssemblyAI6

Get Sentences of a Transcript

Action

In Make“Retrieves all sentences of a transcript.”

Hands back the transcript split sentence by sentence. Shorter units are easier to quote, to count or to send one by one into another tool.

When to use it
to pick out quotes from customer interviews, or to fill one spreadsheet row per sentence for review.
Watch out
look at the output in Run once before adding steps; if each sentence travels on its own, every step on it costs operations.
AssemblyAI7

Get Subtitles for a Transcript

Action

In Make“Retrieves the transcript as SRT or VTT subtitles.”

Exports a finished transcript as a subtitle file, in SRT or VTT, the two formats video players and editing tools read.

When to use it
right after a video interview or a webinar is transcribed, so the subtitle file reaches your video folder without anyone copying timestamps.
Watch out
pick the format your video platform expects before you switch the scenario on.
AssemblyAI8

Make an API Call

Action

In Make“Performs an arbitrary authorized API call.”

Opens the whole AssemblyAI API to your scenario, reusing the same connection. It is the fallback when none of the ready-made modules covers the request you need.

When to use it
for an AssemblyAI feature that has no dedicated module in Make.
Watch out
you write the request yourself, so keep the AssemblyAI API reference open while you set it up.
AssemblyAI9

Purge a LeMUR Request

Action

In Make“Deletes the data for a previously submitted LeMUR request. Response data from the LLM, as well as any context provided in the original request will be removed.”

Erases the data of a past LeMUR request: the model's answer and the context you sent with the prompt both disappear.

When to use it
when the prompt carried client details and your policy says they must not stay stored once the answer is saved elsewhere.
Watch out
like any delete, it cannot be reversed, and Get a LeMUR Response will find nothing afterwards.
AssemblyAI10

Run a Task using LeMUR

Action

In Make“Prompts an LLM with your transcripts using LeMUR.”

Sends your transcripts to a large language model through LeMUR, along with an instruction you write. You get back whatever you asked for: a summary, action items, a draft follow-up email.

When to use it
once a call or meeting is transcribed and the team wants a usable output rather than raw text.
Watch out
test your instruction with Run once on a few real transcripts before trusting it at scale.
AssemblyAI11

Search for Words in a Transcript

Action

In Make“Searches through a completed transcript for a specific set of keywords.”

Looks inside a completed transcript for a set of keywords you choose and reports what it found. Handy when the question is "was this said?" rather than "what was said?".

When to use it
flag support calls that mention a refund, or interviews that name a competitor.
Watch out
the transcript must be finished first, so place this module after the trigger or after a wait.
AssemblyAI12

Transcribe an Audio File

Action

In Make“Transcribes an audio file and waits until the transcript status is "completed" or "error".”

Does the core job: turns an audio file into a transcript and holds the scenario until the result is completed or in error. The next module receives finished text.

When to use it
in most scenarios where the audio and the follow-up live in the same flow.
Watch out
an error status is a result too, so filter on it before sending anything to people.
AssemblyAI13

Upload a File

Action

In Make“Uploads a file to AssemblyAI. The file will be private and removed after transcription.”

Sends a file from another app into AssemblyAI so it can be transcribed. The upload stays private and is removed after the transcription.

When to use it
when your recordings sit in a storage app or come in as email attachments.
Watch out
the Make Free plan handles files up to 5 MB, so test your longest recording before you rely on it.
AssemblyAI14

Wait until Transcript is Ready

Action

In Make“Waits until the transcript status is either "completed" or "error".”

Pauses the scenario until a transcript you already started reaches completed or error. It adds patience to a flow, not transcription.

When to use it
when the transcription was launched by Make an API Call or another path, and later modules need the finished text.
Watch out
Transcribe an Audio File already waits on its own, so adding this after it only repeats the same check.
AssemblyAI15

List Transcripts

Search

In Make“Lists transcripts from your account.”

Returns the transcripts stored in your account, each one as its own bundle. It is a search, so the answer can be zero, one or many results.

When to use it
for a cleanup scenario that deletes old transcripts, or to rebuild a log of everything already transcribed.
Watch out
every result flows into the next modules and each run costs operations, so filter early.
Need help

Need help automating AssemblyAI with Make?

A person reads every message.

FAQ

AssemblyAI and Make: common questions

01Is the AssemblyAI Make integration free?
Yes, on the Make side. AssemblyAI is a standard app, available from the Free plan. That plan allows 2 active scenarios, at least 15 minutes between two scheduled runs, 5 minutes of execution per run, files up to 5 MB and 512 MB of data transfer. For audio, the 5 MB ceiling is the one to watch. Paid plans (Core, Pro, Teams, Enterprise) lower the interval to 1 minute, remove the cap on active scenarios and allow 40 minutes of execution. Make counts your usage in operations, month by month, and AssemblyAI bills its own service separately.
02What do you need to connect AssemblyAI to Make?
An AssemblyAI account and a Make account. The documentation lists no other prerequisite. Add any AssemblyAI module, click Create a connection, give it a name if you like, then either authorize Make on the AssemblyAI page or paste the key your account provides, depending on what the window asks. Click Save. That single connection then serves all 16 modules, so you set it up once per account. If several people share scenarios, a clear connection name avoids mixing accounts.
03Does the AssemblyAI trigger in Make react instantly?
Yes. Watch for Transcript Ready Notification is an instant trigger: AssemblyAI calls Make through a webhook as soon as a transcript ends, either completed or in error. There is no polling interval to set on it, unlike a scheduled trigger that Make checks every 15 minutes by default. If your scenario starts on another app instead, for example a storage folder, the pace depends on that first module. Remember that error notifications arrive too, so add a filter on the status.
04What if a module you need is missing from the AssemblyAI app in Make?
Use Make an API Call. It reaches any endpoint of the AssemblyAI API with the same connection you already created, so a feature without a dedicated module is still within reach. You write the request yourself, which asks for a look at the AssemblyAI API reference. The official documentation publishes no specific limits for this app, so the constraints you will meet first come from your Make plan: file size, execution time and the operations you consume.
05Should you use Make or n8n for AssemblyAI?
It depends mostly on where your other automations already run. AssemblyAI has an app in Make with 16 modules and also a node in n8n. In Make, the AssemblyAI app is a standard app, available from the Free plan, and Make an API Call covers what its modules do not. The AssemblyAI n8n integration page describes the other side. Compare the module list on each side with what your scenario needs, starting with the trigger and LeMUR tasks, before you decide.
Same platform

Other docs you might need on this stack.

Browse all integrations