Deepl and Artificial Intelligence

General discussion about learning languages
Kraut
Black Belt - 2nd Dan
Posts: 2619
Joined: Mon Aug 07, 2017 10:37 pm
Languages: German (N)
French (C)
English (C)
Spanish (A2)
Lithuanian
x 3225

Deepl and Artificial Intelligence

Postby Kraut » Mon Aug 15, 2022 1:54 pm

https://www.tripadvisor.es/ShowUserRevi ... lonia.html

Original:
Llamé para que me trajeran la pizza a casa con servicio a domicilio. Al llegar el chico me dijo que no podía pagar con tarjeta.

Translation into English:
I called to have the pizza delivered to my home. When I arrived the guy told me he couldn't pay by card.

Translation into German (via English)
Ich rief an, um mir die Pizza nach Hause liefern zu lassen. Als ich ankam, sagte man mir, dass ich nicht mit Karte zahlen könne.
1 x

Nogon
Green Belt
Posts: 306
Joined: Sat May 13, 2017 6:21 pm
Languages: German (N), Swedish (C), English (?), French (A2), Esperanto (A2). Reading Danish, Norwegian, Dutch, Afrikaans. Wanting to learn Polish, Yiddish
Language Log: https://forum.language-learners.org/vie ... 15&t=16039
x 1073

Re: Deepl and Artificial Intelligence

Postby Nogon » Mon Aug 15, 2022 3:19 pm

Kraut wrote:Translation into English:
I called to have the pizza delivered to my home. When I arrived the guy told me he couldn't pay by card.

Oh, you get payed when eating pizza? I don't dare to imagine that pizza's quality! :D
3 x
Assimil French : 65 / 113
Active wave : 15 / 113

User avatar
Le Baron
Black Belt - 3rd Dan
Posts: 3578
Joined: Mon Jan 18, 2021 5:14 pm
Location: Koude kikkerland
Languages: English (N), fr, nl, de, eo, Sranantongo,
Maintaining: es, swahili.
Language Log: https://forum.language-learners.org/vie ... 15&t=18796
x 9570

Re: Deepl and Artificial Intelligence

Postby Le Baron » Mon Aug 15, 2022 3:43 pm

Google does better by translating Al llegar as 'upon arriving/arrival'.

They get lost when the his/her + noun is undifferentiated. Also with subjunctives and another complicated constructions. It's to do with the input though. The thing is managed by users who get things wrong.
2 x
Pedantry is properly the over-rating of any kind of knowledge we pretend to.
- Jonathan Swift

Odair
Yellow Belt
Posts: 96
Joined: Mon Feb 07, 2022 11:45 am
Languages: en (C1), fr (B2), es (C1), it (B1), pt_br(N)
x 159

Re: Deepl and Artificial Intelligence

Postby Odair » Mon Aug 15, 2022 5:21 pm

Kraut wrote:Al llegar el chico me dijo que no podía pagar con tarjeta.


The sentence is on itself ambiguous if you don't know the larger context.
0 x

tractor
Green Belt
Posts: 380
Joined: Sat Oct 29, 2016 10:58 am
Location: Norway
Languages: Norwegian (N), English, Spanish, Catalan, French, German, Italian, Latin
x 778

Re: Deepl and Artificial Intelligence

Postby tractor » Mon Aug 15, 2022 5:52 pm

Odair wrote:
Kraut wrote:Al llegar el chico me dijo que no podía pagar con tarjeta.


The sentence is on itself ambiguous if you don't know the larger context.

The larger context being the sentence immediately before it.
0 x

Odair
Yellow Belt
Posts: 96
Joined: Mon Feb 07, 2022 11:45 am
Languages: en (C1), fr (B2), es (C1), it (B1), pt_br(N)
x 159

Re: Deepl and Artificial Intelligence

Postby Odair » Tue Aug 16, 2022 2:04 am

tractor wrote:
Odair wrote:
Kraut wrote:Al llegar el chico me dijo que no podía pagar con tarjeta.


The sentence is on itself ambiguous if you don't know the larger context.

The larger context being the sentence immediately before it.


I meant understanding what a pizza delivery service is all about.
1 x

desafiar
White Belt
Posts: 41
Joined: Sat Apr 30, 2022 5:09 am
Languages: English native, studying Spanish.
Language Log: https://forum.language-learners.org/vie ... 15&t=18131
x 144

Re: Deepl and Artificial Intelligence

Postby desafiar » Tue Aug 16, 2022 3:54 am

Odair wrote:
tractor wrote:
Odair wrote:
Kraut wrote:Al llegar el chico me dijo que no podía pagar con tarjeta.


The sentence is on itself ambiguous if you don't know the larger context.

The larger context being the sentence immediately before it.


I meant understanding what a pizza delivery service is all about.


That’s a good point. It's not enough for the translation service to understand the object action relationship of traer in the first sentence carrying over to the llegar relationship in the second sentence. It must know the concept of pagar as it relates to the consequence of the actions.

Interestingly, Apple Translate gets it correct. I tried both DeepL again and Reverso and both got it wrong. As a Spanish learner, this highlights a challenge for me with more complex structures. I've found that I can get off track in some dialogs where a discussion might include a topic involving a third party. It's not a problem in English due to the usage subject pronouns.
0 x

User avatar
ryanheise
Green Belt
Posts: 459
Joined: Tue Jun 04, 2019 3:13 pm
Location: Australia
Languages: English (N), Japanese (beginner)
x 1681
Contact:

Re: Deepl and Artificial Intelligence

Postby ryanheise » Tue Aug 16, 2022 3:58 am

It looks like it gives you alternative translations:

Image

Google Translate has a similar thing, which I think in the past may have been used to help train and improve their translations. My guess is that it probably got abused so they now require a sign-up and verification process to contribute to improved translations.
1 x

golyplot
Black Belt - 1st Dan
Posts: 1744
Joined: Mon Jan 16, 2017 9:41 pm
Languages: Am. English (N), German, French, ASL (abandoned), Spanish, Dutch, Italian, Japanese (N2)
Language Log: https://forum.language-learners.org/vie ... 15&t=12230
x 3461

Re: Deepl and Artificial Intelligence

Postby golyplot » Tue Aug 16, 2022 4:46 am

If you think this is bad, try Japanese, where so much that English specifies is left to context. Like the subject in only implied by context in most cases, as well as genders. I once tried translating a paragraph of Japanese where Google used three different pronouns (he, she and it) to refer to the same character in the same paragraph! And getting confused about who is doing which action like the OP is the rule, not the exception, when machine translating Japanese, since you have to guess it from context.
1 x


Return to “General Language Discussion”

Who is online

Users browsing this forum: No registered users and 2 guests