---
title: "Speech to text"
video_id: wLBF0uo17qU
url: https://www.youtube.com/watch?v=wLBF0uo17qU
workflow_page: https://workflows-of-ai.com/workflow/speech-text
language: en
captions: generated
---

0:01 hey there um Marcus here I'm just going
0:03 to show a very very easy
0:07 um easy not really workflow I'm just
0:10 going to basically show how you can uh
0:14 how you can create more accessibility on
0:16 your website uh using text to speech uh
0:19 Tool uh in this case 11 lab that has
0:22 really really good quality and this is
0:24 going to be the easiest version of
0:26 creating this there's going to be videos
0:28 in the future where you can create even
0:30 podcasts or uh you know radio jingles
0:34 and a lot cooler stuff and it's
0:36 completely automated but in this case
0:38 we're just going to take the easiest
0:40 Road where you want to create a uh
0:44 basically human voiced MP3 version of
0:48 your
0:49 article uh so that people that may not
0:53 be able to read can listen to your
0:55 article instead now I have copied an
0:58 article here from Yahoo but but I kept
1:00 it very short because it the generation
1:03 in La Labs takes some time uh so and I
1:07 don't want you to sit here and wait on a
1:09 video where I have nothing to say while
1:11 we're waiting and I don't really have
1:13 any editing software so I that would
1:17 just be
1:17 awkward so uh as always this is
1:21 druple uh vanilla Drupal installed it's
1:26 uh the AI interpolator Suite but in this
1:28 case you just need a 11 LS module for
1:31 this um if you want to know how to
1:34 install it set it up there's videos in
1:36 the
1:37 description uh also if you want to know
1:39 more about the interpolator video in the
1:42 description everything is open sourced
1:44 everything is free if you like what you
1:47 see you can just download it afterwards
1:49 and set it up yourself of course 11 Labs
1:52 is a service that cost money
1:54 then uh so let's get started we log in
1:57 as a admin or anyone with access to
2:00 change uh field structures we go to our
2:04 content type article go to the fields
2:06 and we add a new field that is the 11
2:09 Labs field and this is going to be audio
2:12 version we call this right so this is
2:14 the audio version of our uh article we
2:19 just need one MP3
2:21 file uh and we can save it like this
2:24 we're going to go to manage display as
2:26 well so uh the 11 Labs module comes with
2:29 something called like a virtual player
2:31 which is one player that's playing
2:33 multiple MP3s since if you create a
2:37 dialog this will create multiple MP3s
2:40 but you might want to play this so it
2:42 looks in the front end like it's just
2:44 one but in this case we just have one
2:47 file so we can use a normal audio
2:49 element uh that's easier for
2:53 accessibility reasons uh we save that go
2:56 to our
2:57 content and then basically basically the
3:00 only thing we have to do is copy the
3:01 article into the text of the audio
3:04 version select a voice and these are the
3:06 voices you have in your 11 Labs account
3:09 so you see I have my own voice here for
3:12 instance but we're not going to use that
3:15 in this case what we're going to use
3:18 is uh let's see let's use the Freeman
3:24 voice so either you can click generate
3:26 audio so you can preview it before you
3:28 save it or you can click save it gets uh
3:30 generated I click on the generate audio
3:32 so we can preview it before um
3:37 saving oh and there was one thing more I
3:40 should have done I can do that in a new
3:42 tab so if we go back to our content tab
3:45 we
3:47 want the audio version to show up before
3:50 the actual article as
3:53 well so let's see here if we can get the
3:57 audio file so it it does take some time
4:00 to generate the audio there are also I
4:03 can show you while this is happening if
4:05 we go to content type article and we
4:08 manage the form display so that's what
4:10 you see here
4:12 basically uh in the audio
4:16 field you can basically show Advanced
4:18 field and there's some knobs some
4:21 sliders that you can tweak there so you
4:24 can make it faster or make it slower and
4:26 more realistic so now you see it's
4:28 finished here
4:30 so we can play a little bit Apple's
4:32 iPhone 15 Pro Max was supposed to take
4:35 the World by storm so you hear it there
4:37 but we save first so we can see it in
4:39 the article we go to the article and we
4:42 have the audio version here we can go
4:44 seem to solve its thermal problems it
4:46 came at the expense of performance as
4:48 a17 benchmarks dipped following the
4:51 update this is a very narrative voice so
4:53 it's actually horrible for article was
4:56 not a good choice but you get the idea
4:59 right
5:00 um so that's how you can use AI to
5:03 create audio on your
5:06 articles thank you for watching