Skip to main content

Trying out ScribeTube

Since writing this post, the Scribetube site has now disappeared (checked 26.6.25).Other free transcription sites are available.

………………………………

Larry Ferlazzo on BlueSky often shares AI tools he has come across. ScribeTube is one such tool. The idea is very simple. You go to the ScribeTube site, enter the URL of the YouTube video, select the language from a comprehensive list and it instantly creates a transcript of the video.

I tried it out with four videos in French to test its effectiveness. The first video is this one - a female speaker describing her weekend at level A2 in pretty clear and slowly delivered French.


Here is an extract of the transcript:

Bonjour à tous ! Aujourd'hui je vais vous parler de mon week-end à deux vitesses différentes. Je vais commencer par parler lentement pour que vous compreniez bien tous les mots comme d'habitude, mais dans la deuxième partie de la vidéo, je vais accélérer et parler de façon un peu plus naturelle, comme je le ferais avec un autre français natif.  C'est parti !

Alors, j'ai passé un très bon week-end, le samedi je suis restée tranquille à la maison, je me suis reposée, j'ai fait une petite sieste dans l'après-midi, mais j'ai aussi rangé la maison, trié des affaires, enfin bref, les petites choses du quotidien qu'il faut faire souvent le week-end!

Mais, en fin de journée, le samedi, je suis allée boire un verre avec des amis.  C'était très agréable, il faisait beau, on a bu un verre en terrasse à l'extérieur.   

I had to fix a few gapping errors and the occasional punctuation error, notably hyphens.

Next, here is a video from the Portrait d'enfant collection by Arte. I really like these for level A2 listening (GCSE). Eye-opening and at a reasonable level for Higher GCSE (A2).


Here is what ScribeTube produced (bear in mind that there is background music and chat)

Bonjour je m'appelle C j'ai 8 ans et je vais vous montrer ma maison je suis ravie de vous accueillir suivez-moi voici ma mère mon frère ma tante nous vivons avec mes deux oncles et tous mes cousins et voilà mon père il porte le turban traditionnel du Rajastan venez je vais vous montrer ma chambre je dors ici avec mes parents c'est mon lit là il y a une photo de notre temple devant laquelle je pris tous les [Musique] jours nous avons TR cobras quand je touche le serpent il est froid Frid et ils perdent leur peau tous les mois celui-ci fait 2 m j'aime le mettre autour de mon coup ici tous les enfants aiment beaucoup jouer avec eux nous n'avons pas peur car le venin était retiré je vais les nourrir on le donne du poulet le matin et le [Musique] [Musique] soir dans ma famille nous sommes charmeurs de serpent depuis TR génération en Inde les filles ne sont pas censé charmer les mais Mo j'ai envie de faire mon père commence à m'apprendre tu te places sur le 4 5 6 et 7 et après tu souffles les serpents n'entendent pas la musique ils ressentent seulement les vibrations car ils n'ont pas d'oreille...tu dois faire des mouvements de gauche à droite pour qu'il suivent l'instrument [Musique] seant j'étais contente de partager ce moment avec vous merci et [Musique]

In this case there are many full stops missing and a few issues I have highlighted in bold, but overall, a very good transcript with little editing needed.

Now let's try a higher level. This is a video I created a resource for on my site. It's an analysis of the popular A-level film La Haine



This was a fail. I got the message: Automatic subtitles not found in YouTube, try manually generating using filetranscribe.com. I went to that site, but you need to upload an MP4 file, for example, so you would need to download the YouTube video first. There are ways of doing this, but I did not explore them even though I have a YouTube premium account. I wanted to keep this fast and simple.

Lastly, I went for an A-level video with closed captions/subtitles. I have a resource to go with this on my site.


This time the transcript was excellent. Here is an extract:

Je réalise des émissions de flux, c'est à dire des magazines, des documentaires, des divertissements. Le réalisateur, sur le tournage, c'est un peu comme un chef d'orchestre en musique. C'est lui qui va donner les consignes aux cadreurs, au son, à toute l'équipe technique. Souvent en France, un réalisateur est un auteur réalisateur. Souvent, c'est quelqu'un qui est partie prenante dans l'écriture du scénario. Il s'occupe de prendre ce scénario et de le transformer pour faire en sorte que le film devienne réel. Il va définir comment on va filmer l'émission, où on met les caméras, comment on éclaire. Il supervise aussi bien le tournage, que le montage et la post-production. Il faut s'intéresser à l'être humain, il faut être sensible aux gens et leur raconter leurs histoires en fait.


So overall, provided your source video has the letters CC highlighted, it looks like ScribeTube does a very good job of producing a trasncript which you could use in various ways, whether it be for gap-fill, questions (e.g. produced with Chat GPT), false sentences to correct, true/false or a faulty transcript task (where you gibe students a version of the transcript with a number of errors to find by listening to the video).

As is usually the case, the tool does not do anything you couldn't manage yourslef, but it does it instantly. That said, teachers with weaker transcription skills would find it particularly useful, I imagine. Do let me know if I have missed anything obvious here or any other uses you have found for ScribeTube or other similar tools.








Comments

Popular posts from this blog

What is skill acquisition theory?

For this post, I am drawing on a section from the excellent book by Rod Ellis and Natsuko Shintani called Exploring Language Pedagogy through Second Language Acquisition Research (Routledge, 2014). Skill acquisition is one of several competing theories of how we learn new languages. It’s a theory based on the idea that skilled behaviour in any area can become routinised and even automatic under certain conditions through repeated pairing of stimuli and responses. When put like that, it looks a bit like the behaviourist view of stimulus-response learning which went out of fashion from the late 1950s. Skill acquisition draws on John Anderson’s ACT theory, which he called a cognitivist stimulus-response theory. ACT stands for Adaptive Control of Thought.  ACT theory distinguishes declarative knowledge (knowledge of facts and concepts, such as the fact that adjectives agree) from procedural knowledge (knowing how to do things in certain situations, such as understand and speak a langua...

Zaz - Si jamais j'oublie

My wife and I often listen to Radio Paradise, a listener-supported, ad-free radio station from California. They've been playing this song by Zaz recently. I like it and maybe your students would too. I shouldn't really  reproduce the lyrics here for copyright reasons, but I am going to translate them (with the help of another video). You could copy and paste this translation and set it for classwork (not homework, I suggest, since students could just go and find the lyrics online). The song was released in 2015 and gotr to number 11 in the French charts - only number 11! Here we go: Remind me of the day and the year Remind me of the weather And if I've forgotten, you can shake me And if I want to take myself away Lock me up and throw away the key With pricks of memory Tell me what my name is If I ever forget the nights I spent, the guitars, the cries Remind me who I am, why I am alive If I ever forget, if I ever take to my heels If one day I run away Remind me who I am, wha...

What is the phonological loop?

This post is about how we use part of our short-term memory (working memory) to process sounds, words and longer utterances. I also intend to show how knowing about the phonological loop can help you refine your practice as a language teacher. Firstly, what is the phonological loop and where does it fit into a popular model of working memory? To start with, it's probably best to start by activating another component of short-term memory, your visuo-spatial sketchpad. Look at this diagram: Image from cheese360 at Wikimedia Commons That is one depiction of the well-known model of working memory put forward by cognitive psychologists Alan Baddeley and Graham Hitch back in 1974. But first, when we see, hear, touch, taste or smell something our sensory memory takes note (beneath our consciousness). As far as language is concerned, we choose to pay attention to it and the information enters working memory, more specifically what are called the visuo-spatial sketchpad (aka scr...