---
title: "Transcribe audio and summarize"
video_id: ifXQDbMQMJY
url: https://www.youtube.com/watch?v=ifXQDbMQMJY
workflow_page: https://workflows-of-ai.com/workflow/transcribe-audio-and-summarize
language: en
captions: generated
---

0:01 hey everyone uh Marcus here again with
0:04 another workflow uh in this workflow I'm
0:07 going to introduce the deam module so
0:09 deogam is a a service that can basically
0:13 take a audio file uh and transcribe it
0:16 for you in text so what we're going to
0:18 do today is we're going to take we're
0:21 going to set up so we can summarize uh
0:25 uh a MP3 file basically what's being
0:28 said in MP3 file
0:31 uh as usual what you see in front of you
0:33 is uh Drupal with the AI interpolator
0:36 modules installed
0:38 um everything is open source everything
0:40 is free you're free to use it after you
0:43 have looked at this um in the future
0:46 there's going to be in the video
0:48 description
0:49 some links to videos that uh we show you
0:53 how to set up drle as well so off we go
0:57 we start by logging in and we're just
0:59 going to create that entity type for
1:01 this as always it doesn't have to be
1:06 nodes uh usually you work with custom
1:09 entities but in this case we just do it
1:12 so I can showcase everything over the
1:14 GUI
1:16 uh uh what do we call it summary of
1:21 audio
1:23 file and we just save
1:27 this we don't need the body field that
1:30 comes with Drupal by
1:32 default um so the first thing we're
1:35 going to do is we're going to create a
1:36 file
1:37 field that is the audio right so we're
1:41 going to upload the audio file into
1:43 there it's just going to be one you
1:45 could do multiple and it would create
1:47 multiple text fields of course only
1:50 allowed MP3 you can do wob files as well
1:53 you can do AAC files as well but in this
1:55 case we take
1:57 MP3s um and then then we're going to
2:00 just add a text plane
2:03 long that we call
2:07 transcription so this is where the
2:09 transcription will end up after we have
2:12 uh uploaded right and for this to happen
2:14 I have to enable the A interpolator and
2:17 choose the rule uh deep gr
2:20 transcribe and we base this on the audio
2:23 right I'm going to set up so it's
2:26 running in batch mode and Save so now
2:30 it's going to take that audio and
2:32 transcribe it to this field but we want
2:34 to take it one more step we want to
2:35 summarize it right so we create another
2:38 text long that we call
2:43 summary um and that's just just be one
2:46 field as well once again enable the AI
2:50 interpolator choose the open AI string
2:52 long so we're going to take a text field
2:55 and do a prompt on how to create uh
2:58 another text field right and we're
3:01 basing this on the transcription So
3:03 based on the
3:08 transcription
3:10 create a two sentence summary
3:15 of what the person talking
3:21 about uh what the person
3:24 talking is
3:27 thinking so what I'm going to use
3:30 uh
3:31 is uh
3:35 MP3 uh that comes from uh Daman Le is
3:40 talking uh
3:42 about
3:44 um asking leer
3:47 basically so anyway we need to set the
3:51 context so the transcription and the
3:53 context then comes from transcription
3:56 here right from that field and here it's
3:58 important that we first of all do batch
4:00 mode but we also set the higher
4:03 interpolation weight so the
4:06 transcription has to exist before we do
4:08 the summary
4:10 right and that's it then we just add a
4:14 common uh content that's called summary
4:16 of audio file I have to give a title
4:20 here uh and then we're just going to
4:22 take this
4:24 here uh oh it's from Julia Caesar sorry
4:28 so that's what we're going to do and
4:30 this is then going to fill in first
4:32 transcription with what uh Daman Lewis
4:34 is saying in this piece and then a
4:38 summary basically of what he
4:41 said so we click
4:45 save and first it uploads creates a
4:49 transcribe and then it summary so we
4:52 have here transcriptions from what he's
4:54 saying right and then a summary the
4:58 person thinking about the funeral say
5:00 Cesar and the conflicting options
5:02 opinions of his
5:05 ambition um and that's basically it um
5:10 it's very very simple but opens up a lot
5:13 of possibilities where you can uh not
5:17 just upload MP3 files because of course
5:20 if you're on web you can uh you can of
5:23 course allow the person to speak into
5:25 the microphone for instance so if you
5:28 want to make one of these um startups
5:33 that I have seen that you know
5:35 summarizes your ideas from talking you
5:38 could do it with this so that's it thank
5:42 you