Looking for European language Permies that have made cheese! Proofreading a book translation.

I had an idea of translating my cheese book into different languages. There’s an AI thing that is good at translating, but because I am not familiar with cheesemaking terms in other languages, I am not confident to put out a book unless I know that a human that has made cheese before (or has read cheese books in the language and knows all the terms) has looked over it, and corrected any mistranslations.

For example, I am not sure if the terms I use, such as culturing and rennetting, would translate correctly. The more standard terms might be “ripening” and “coagulation”, but even with these I am not sure how they would translate.

The work would involve reading through the translated book, cover to cover, and if any translated term jumps out to you as mistranslated, or there are any other errors, to let me know. The book is around 66,000 words.

This is paid work. Probably around 250 euro per language.

It would be as part of a stretch goal for my cheese book Kickstarter here: https://www.kickstarter.com/projects/706848724/make-your-own-cheese-natural-small-batch-cheesemaking-book?ref=1nzyel

Languages I am considering translating into:
German
French
Italian
Spanish
Portuguese
Maybe more languages if there is enough demand for it, and if there is someone to help me.

Hi,

If I may: I have been asked to translate from English into my two first languages several time, and sometimes to ""just" proofread something from AI".
I would never ever accept only 250$, to check/translate over a 100 pages book. Checking seriously takes more in the range of hours per page of text than in the range of minutes.

Furthermore, when asked to translate, I rather have the original text than a text already gone through AI, as I also have access to these tools.

hans muster wrote:Hi,

If I may: I have been asked to translate from English into my two first languages several time, and sometimes to ""just" proofread something from AI".
I would never ever accept only 250$, to check/translate over a 100 pages book. Checking seriously takes more in the range of hours per page of text than in the range of minutes.

Furthermore, when asked to translate, I rather have the original text than a text already gone through AI, as I also have access to these tools.

Thank you for the feedback. I had no idea it would take so long! I've proofread stuff in English before and can do it quite quickly, so had thought it would be similar for translations.

Kate Downham wrote:...

Languages I am considering translating into:
German
French
Italian
Spanish
Portuguese
Maybe more languages if there is enough demand for it, and if there is someone to help me.

Hi Kate,
if you decide to have it translated also in Dutch (the language of the Netherlands, not very different from Flemish, the language in the North of Belgium), maybe I can help you.
I am not looking for a paid job, but I would voluntarily (and without obligation) try to read the translation (made by AI), see if I understand the meaning, ask questions (through e-mail) if something isn't clear to me and it might be a translation error.

Added /edited: I don't have any experience in cheese-making ...  
-Inge-

Reply to Kate:

do you have some interest in Spanish language assistance?

I am willing to help more, but not so very keen on public posts like this:  The other posts seem so confident, so convinced of how things are.


Let me know if you wish to discuss Spanish more...

I may be able to help with German. I live in the US though.

Although my first inclination is to agree with Hans, I think professional translators have a different relationship with their work and our professional profile (and being responsible for risk if there are mistakes, reputation, we have specific software and tools for quality control that take years to learn...). We do this every day and talk to our clients about why machine translation and AI can be harmful for them, and so it's almost a sore spot.

That said, there's no reason you can't do exactly what you propose and get a great outcome. Not everyone who grows tomatoes has to be a large-scale professional farmer, you can plant one in your yard and get just as good tomatoes, I think here it's the same. Your volunteer or semi-volunteer staff can do just as well, not every job has to be a professional production.

If I can offer a bit of advice as someone who does this for a living: what's more important is that the readers 1) understand cheesemaking, more than anything else, in both languages and 2) can be honest about whether something "sounds funny" to native speakers of the target language. Maybe also keep an eye out for things that can be tricky for people moving between English and your language (if you're a native speaker of a different language but are here writing in English on this forum, you are probably familiar with several of these!)
Another suggestion is to have two people look at each language, if possible, just to make sure. Maybe have the second one read over the first one's work. The AI will make some crazy mistakes, it's basically guaranteed, and two pairs of eyes will make sure that your final product is better. Even if you end up having to hire someone on a low-cost platform like Fiver, just for a quick proofread. But make sure the first person who does the real work is the one who understands cheese.

My Portuguese is Brazilian, not European, so I'll leave that for someone else.

Wanted to add: you bring up a good point about whether the vocabulary will translate. Words can always be translated, but the concepts and products may not. This is why it's a good idea to have a person who understands cheese and milk or at least buying cheese and milk involved. In translation we talk about 'transcreation', when you have to not just translate but also add information. For example, if I'm translating from Japanese and there's an important cultural thing (like different words used by older/younger people) I may need to add some extra information so readers can understand. Here I can only get liquid rennet, maybe in Italy they only have powdered rennet, (no idea if this is true, but just as an example): this would be really important information that your italian editor can help you with. This is why translators charge a lot.

Hi, Kate!

First want to say how excited I am to hear of your cheesemaking book. I wish you great success on your Kickstarter.

I really hope you opt for professional, HUMAN translators instead of AI. We're in touch by email, so reach out if you'd like to get an estimate from a few localization studios or even individual folks who can translate English to Japanese, Spanish, and other languages. I know of many of them in the game industry as a regular part of my work work there designing and writing game stories.

Best of luck!

I usually just lurk here, but I must speak up about this,

As a translator by trade (Spanish), I want to take every AI out there and smash it to bits with a sledgehammer. They take work away from us who make a living from translation, editing, and proofreading,

Yes, Google Translate can be very helpful but it will never capture nuance the way a human being can. I feel like it's more work to correct it than to translate directly.

Also, current AI models run on what amounts to plagiarism, they are fed content from living authors who are neither acknowledged nor paid for the use of their work.

I am willing to help translate and proofread for free as part of a community effort, as long as said effort is acknowledged in the final product (such as listing the members of the proofreading/translation/editing team at either end of the book).

I have to agree totally to the comments of the Pro's.
It is a totally different approach to translate a novel or poems in another language and still have the spirit of the original alive.
Also AI translations are often false, if you have content, which is not so common.

I am a hobby cheesemaker and my native language is German - and as this is more a technical manual (no offense) I would be willing to volontier.
I would say however that there should be two or three people translating: we could split the content into several parts, translate it and proof-read the translated text of the other people.

Payment: Why not offering something from permies for free??

Andreas

Thank you all for your feedback.

I am a big fan of the first two Terminator films, and wish I could smash AI into many pieces as well. As well as dystopian possibilities, I feel for those who are losing work because of it.

On the other hand, it is there, it’s not going away, and from what my husband and I have seen, it is actually reasonably good at translating English to German. I don’t think I’m aiding Skynet by feeding it information about making cheese.

I also don’t need the most beautiful translation possible - I am just looking into the idea of translating it so that it can be understood and people can make cheese from it in other languages.

The reason I am considering using AI is for budget reasons. I have put hundreds (maybe thousands - I don’t count) of hours into creating this book. When my Kickstarter hit 100% funded, I had made $0. Now it’s gone beyond 100%, I am starting to get paid, but I’m not in a position to pay anyone an hourly wage for translation. If I were to do that, I’d have to do an entirely new Kickstarter with maybe 20x the funding goal of the first one.

For stretch goals, I can afford to budget a lump sum (the 250 euro mentioned above), put up an amount to reach on Kickstarter that will leave me with that much spare after Kickstarter expenses, and then I’d be able to do it. The translation idea will also involve more unpaid hours from me, so I am not sure if I want to take on that extra work either, as it’s hard to say whether people would buy this in other languages or not.

The best thing might be for me to wait until the Kickstarter is over and books are all sent out, and then have another think about it.

As far as quality goes, the AI program I’m familiar with, from what I have seen in German translations, is way superior to Google Translate or any of the klunky older translating things. I might try translating a recipe into a few of the languages mentioned here, and post in this thread.

Kate, I hear where you're coming from. Every client has different financial resources and abilities, and part of being a professional in this area is understanding that and offering options.

I occasionally get clients who want books translated, and before I hit them between the eyes with a 5-figure fee, my first suggestion is always to wait til the first edition goes out (as you've noted above) and then talk to your publisher. They often have in-house services and can do a better job for free. It also saves you the hassle of choosing what languages you want to do: they will have a better view of the market and where you can expect to sell.

I don't remember if you're self publishing; if you are and you don't have this resource, then that bit of time will give you an idea of where your efforts are best spent. I'd take one language at a time, maybe after you get some requests. And maybe if you have a place where you're promoting your book, see if there is some way to capture requests for translation so you can analyze and know for sure whether to start with Dutch or Italian or whatever.

And while I'm glad Skynet will not be killing us with cheese, the biggest risk for you as an author (using AI services) is that it will end up in some sort of auto-harvesting thing that produces fake books online for sale with your name, ostensibly this book (maybe with only a few bits of your content). If I were you, I would set a google alert or use some sort of other automated tool for your name and the title of your book just to keep tabs on what's going on on the web.

Edited (again) to add: some years ago I was involved in a similar project for a vegan cookbook. the author realized, as you did, that she didn't have the funds to pay for it. so it became a crowdsourced thing. everyone who was involved was a volunteer, and got listed in the book and a free copy. i don't believe in telling people to do their job for free "for exposure", which is almost always a load of bollocks, but doing a job for love, if you love the project, can be a different story. at least you're being upfront about the situation, i think it is a bit better than a token payment. if i had the resources to make cheese i would be all over helping with this.

Hi Kate,

I'm a Spanish (European) native with a linguistic education and I've worked as an English > Spanish translator for a few years.

A typical rate for English > Spanish translation is 8 cents per word. For 66.000 words, that's 5.280 Euros. The lowest I've ever gone, only for projects I was passionate about, was 4 cents per word. For 66.000 words, that's 2.640 Euros. It's ten times the amount you were planning per language, but I could do it for that.

I'm working in a different field now and sort of missing translation. I find myself with lots of free time right now and in need of a few extra bucks, so I would be willing to possibly halve my lowest ever rate for this one project. That'd be 2 cents per word, so about 1.300 Euros (and then I still have to pay taxes...). I would need to see a few pages worth of text first before making this offer, though, to judge the difficulty of the translation.
I understand that's probably still way too much money for just one language, but it's a ridiculously low rate and sadly I just can't afford to work for less. 66.000 words is a full month worth of full time work - and that's just for translating! Then comes proofreading and editing, which can easily take half as much time again or even more.

If you would like to consider my offer or negotiate, you can contact me per email (sent you a private message), gladly also in the future. Otherwise, I'd suggest you to see if someone is willing to do it just purely out of love - there's always some, and amateur quality can still be worth it when it lacks a price tag. Definitely do not use AI for the translation: the technology is good enough to trick you into thinking it's doing a good enough job, but it's not far enough yet to *actually* do well with a longer text. Considering the nature of the project, some incorrect or inconsistent translation here or there could very easily ruin entire recipes. AI powered translation is a fine tool in a good translator's toolbox (despite the luddites claiming that it's the end of their careers), and it's a good tool in a pinch for the common user, but it really isn't something you should use to do the bulk of the work if you care at all for clarity and accuracy.

Another option is contacting translation bureaus. They usually offer heavily discounted prices if you get multiple languages at once - although you'll likely still be looking at a 5 figure price tag anyways, and in my experience the quality is often rather serviceable at best.

Either way - congratulations on the successful Kickstarter! I'm looking forwards to my copy

Hi Kate!

As a professional translator, I have noticed that AI sometimes does a pretty good translation job. I found that it mostly depends on the language combination: the more common the languages, the better AI does. That, in combination with the online presence of the languages and the complexity of the languages. For example: an AI translation English - Spanish will have a better result than a Dutch - Croatian translation.

To get the best result when using AI for a translation project, you want to involve two people to process the AI translation:

The first person to check a translation is called a revisor. This person will absolutely need the source text! It is someone who understands both the source language (the original text) and the target language (the translation), and has knowledge and thorough understanding of the cultures, translation strategies, localization, etc. This is very important, because that knowledge is exactly what AI does not have (up untill now) and where it will mess up badly!

The second person involved in the translation process is the proofreader. This person is a native speaker of the target language, reads the revision and does not need the source text. A proofreader has the sharpest eye for detail and will discover sentences that sound a little strange, will find typos, inconsitencies, etc.

Skip the revisor (person 1) and you'll end up with a beautifully polished text, but where AI messed up, (and it certainly will, many times!) it won't be caught because the proofreader does not have acces to the original and thus can not know the original content!
Skip the proofreader (person 2) and there will be less (hopefully none) mistakes contentwise, but almost certainly there will be some typos, or some non-native sounding sentences. A revisor's job is often more difficult than a translator's job!

I understand that you propose to skip the proofreader, which is fine, I just wanted to explain a little bit that who you are looking for then is called a revisor, and that person will need the source text to be able to get the content across correctly, the content that you wrote, which is what this is all about :)

If you consider having your book translated into Dutch and need a human being to do the revision, I'm willing to help. I have milked goats, have made cheese a few times (loved it!) and I have used whey on my earthen floor. So, I am not an expert, but I would love to help!
Prices and payments can be talked about when the time is right. Talking politics, I keep for when we meet in the bar for a drink :D
Anyway, there are so many possibilities to come to an agreement :)

For now, I wish you all the luck with the original book. You have put so much work into it! Congratulations!

@Kate Downham Thank you all for your feedback.  For stretch goals, I can afford to budget a lump sum (the 250 euro mentioned above)

Kate, did I get mixed up?  You need this in Spanish and am going with AI?

I am likely "confused" of the post you already made:  Did I ask already if the market is Spain Spanish or Mexico Spanish?


Working with native Mexico speakers directly, it is possible to get very close to that payment and the people will happy to do the work and will do a great job.


At least that is what I experience when I pay to have two projects, books, including a complex one.

Best wishes of your project.  It is certain to work out fine!!

Anne Cummings wrote:@Kate Downham Thank you all for your feedback.  For stretch goals, I can afford to budget a lump sum (the 250 euro mentioned above)

Kate, did I get mixed up?  You need this in Spanish and am going with AI?

I am likely "confused" of the post you already made:  Did I ask already if the market is Spain Spanish or Mexico Spanish?


Working with native Mexico speakers directly, it is possible to get very close to that payment and the people will happy to do the work and will do a great job.


At least that is what I experience when I pay to have two projects, books, including a complex one.

Best wishes of your project.  It is certain to work out fine!!

Hi Anne,

I just got too overwhelmed with finishing up the English version and the Kickstarter - I am definitely still contemplating the idea of doing translations in future, but at the moment I think I'll just focus on getting the original book out, and might come back to the translation idea later.

That is an interesting point you make about European Spanish or Mexican Spanish - I know there are some differences between the two but it's something I hadn't considered when I first thought of this idea.