BookTitres.com - learn foreign languages ​​in the original, listening and reading

Many use subtitled films to listen to speech and read at the same time. I want to offer an alternative - listen to audio books and read their text with interlinear translation and access to each individual phrase. They are available online at BookTitres.com .

The most interesting

You can find books already available for reading in the section “ Listen and read ” on the website. The Hobbit book is over - it's 11 hours of audio and 13,800 phrases. After completing it completely, you will feel a noticeable improvement in your English. Especially if you repeat each phrase after the reader. From the other books basically the first few chapters have been made. If you want to continue - answer the survey and indicate the desired book.

I do not so much pursue the goals of PR, I want to talk about the idea itself and ask for advice on how to develop it. You can leave your wishes by looking at the site at the links above. And if you are interested in reading about the idea and its technical implementation, then look under the cat.

At the end of the article are questions that I would really like to receive answers from the Habr’s people.

How did the idea come about

I always loved to read books and felt a craving for foreign languages, but for a number of reasons I was limited in my ability to learn and practice them. There was not enough patience to study textbooks. I wanted to dive into the environment, reading foreign books in the original. I tried to read a book and a translation in parallel. Then I got acquainted with the method of Ilya Frank, where each phrase of the original is followed by a literal translation in brackets. But there have always been doubts - do I pronounce the word correctly? Often looked in the dictionary for transcription, and it exhausted.

As for the sound, separately without text, it was difficult to perceive, since the main channel of perception is visual, through reading.

Like when text and sound were together. Such a bunch could be found in films with subtitles in the original language, or by downloading a book - audio and text version. But then the following difficulties arose:

• combine the texts of the original and translation;
• combine texts with audio - sometimes a little distracted, and you lose the "thread"
• it is problematic to rewind to the phrase that you did not manage to hear or want to listen to several times in a row; you always catch too much when rewinding;

In general, it took a lot of time and effort to do this. Therefore, the idea came up, to create a format in which each phrase would be associated with a link: a
piece of audio - the text of the original - the text of the translation

and then you could return to them as many times as you like. How to use it in the end?

Made 2 modes: player and text .

In the player, you can turn on all audio and watch text as subtitles, or switch between phrases. And in text mode, you can scroll through the entire text and, if necessary, click on phrases and listen to them.

Since everything is divided into phrases, I added a dictation mode , in which you can set the number of repetitions of each phrase and the delay in front of them.
Many repetitions and a short pause - it is convenient to memorize the material ("zombie" while reading and listening many times in a row).
A big pause and few repetitions - you can write a dictation (you listen to the phrase, and while the pause lasts - you write it, then the next one, etc.)

 

Matter of technology

In the end, it came down to creating 3 text blocks.

1. seconds of the beginning and end of a phrase in an audio file
2. original text
3. translation text

And, of course, specify a link to the audio. What is shown in the screenshot, the beginning of Dickens’s book Great Expectations, looks like this in the code:

0.388889 2.000000 1
2.484127 3.911565 2
4.290023 5.198639 3
8.081406 10.267120 4
10.396145 12.082993 5
12.421542 16.692517 6
16.861451 17.271202 7


Great Expectations
by Charles Dickens
Chapter 1
My father's family name being Pirrip,
and my Christian name Philip,
my infant tongue could make of both names nothing longer or more explicit than
Pip.


Большие надежды
Чарльз Диккенс
Глава I
Фамилия моего отца была Пиррип,
мне дали при крещении имя Филип,
а так как из того и другого мой младенческий язык не мог слепить ничего более внятного,
чем Пип,


 

Public script

All the scripts I have written are available for download in the Create Books section . Guides, updates and changes will be published there. In addition to the front-end (book player), there is also an editor mode.

The end result is an html file, simple in structure, which stores the breakdown of the book by phrase: their seconds and texts in two languages. You can run the book file in almost all browsers, including mobile. Also, if you upload book files on Android, you can run the book from your phone. There is a mobile version of the display and desktop.

 

The working process

The essence of the process is described briefly, if you want to repeat the process, see the video guide . Only the process of obtaining seconds of phrases in AEGISUB by manual selection can be replaced by the automatic one obtained in Audacity. 

 

1. timing phrases

Program Audacity comprises function Sound Finder, automatically vydyalyayuschuyu timeslots all sentences in the audio file. If you want to work with your hands and for sure, you can use the AEGISUB subtitle program .

 

2. Aligning text with audio

When the seconds are received, on a special page they turn into squares with numbers, by clicking which you hear phrases. And you build texts on them. First the original, then the translation.



Creating book titles is much easier than regular subtitles, because:

1. there are almost no extraneous noise in audio books, only speech, and you can recognize the intervals of phrases by program methods
2. the text of the book is in advance, it remains only to transfer lines by audio phrases
3. the translation is also in advance, and often it repeats the original in the order of the phrases

A special case of work when a “book by the method of Ilya Frank” is taken as the source of the text, and the timing is compiled manually in the AEGISUB program, you can watch in this video tutorial: “ Create your books ”.

 

Main difficulty

The selection of phrases in audio is done simply, automatically (Audacity -> Analys -> Sound Finder), the original text in any language can be easily distributed by audio, even having zero knowledge of the language (if you are familiar with writing, of course), you just need to listen to the audio places, press Enter, transferring the text to new lines.

But with the preparation of the translation is more difficult. When you have language skills (I have English, for example), this is easy to do with some translations of books.

Of the 5 English books that have been added on the site, 4 translations follow in the wake of the original, in phrasal terms. And so the work also comes down to wrapping lines in the right places, pressing Enter. (These are Great Expectations , Pride and Prejudice , Jane Eyre andAlice in Wonderland "). Sometimes, in 1 out of 50-100 cases, you have to swap phrases in places.

1 book (" The Life of Benjamin Franklin ") does not match the phrases, it essentially has to be re-translated - but before your eyes there are always semantic translation, it greatly facilitates the work.

 

Breakdown

In my opinion, it’s convenient to split audio files into 5-minute passages. In different languages, this passage contains an average of 100 phrases and takes about 2 mb. Such a portion is convenient for the student to perceive and for downloading to the device with almost any Internet connection. In the work, when creating book titles, this size of parts is also convenient. 

 

Cooperation. Copyright holders

I couldn’t get in touch with Ilya Frank (it seems that he very rarely looks at the mail), so I decided not to abuse his work and made demo versions of available excerpts from books in different languages ​​that I don’t own. Of course, I would like to use the potential of their adapted books by breathing in them the same soundtrack. Of course, by mutual agreement, observing the interests of their publisher.

In the meantime, I decided to focus on books that are in the public domain (whose authors died 70 years ago and earlier). Most books on the site are just that.  

As for modern books, in Western law, especially in English-speaking countries, I found such a thing as "fair use" when you use someone else's content in a modified form, which does not interfere with the copyright holders in the end. It seems to me that it can be applied to our case, since audio books in unfamiliar languages ​​are of little use to people - a very high level of training is required to use them. And if you equip them with pauses (broken down by phrases) and related texts, then interest in them will increase significantly, a foreign-language audience will be added.

It also inspired me that youtube has published audio versions of many popular books with hundreds of thousands of views. Apparently they are not being blocked ... I don’t think that we would really bring damage to anyone. And attention to their products to a wider audience and additional deductions is quite possible.

I would like to hear your advice on cooperation with modern copyright holders of books: Harry Potter, Lord of the Rings, Game of Thrones, House of Cards - called the first thing that came to mind. How realistic is it to interest them? In any case, you can start using the torrent system, where everyone can post anything, but at the request of the copyright holders, this is immediately removed. 

 

Cooperation. Who will make the books? 

Many people translate their favorite poems, songs, create subtitles, scan and recognize books - purely for fun. Conscientious teachers often prepare improvised educational materials for students, investing their own work in this. 

I think all these people might be interested in the format of book titles. They could combine the texts of their favorite work with audio and translation, publish on the site and share with friends, students or simply “fans”.

Recently, one sports judge expressed a desire to prepare international refereeing rules in English in such a format that later taking advantage of the program’s opportunities to memorize them in English and use them to pass the exam in judicial practice.

If you convert popular textbooks into such a format, I think it would be of great benefit to students and the popularity of the service (if you find something to interest the copyright holders). 

You can probably make money deductions and various bonuses for the uploader, depending on its popularity.
My ideal is to make the user-friendly interface for creating, storing and publishing books in online, through the site. Where anyone can post their audio and texts. At the moment, it’s enough to add a simple code with seconds of phrases, texts and a link to audio, and it is displayed in the form of a player and text with phrasal navigation. But registration is not open yet. 

 

Collaboration

Please help me by writing your advice on:

1. Design and usability of the program and site
2. How to design the script source (add to github, provide the code with comments, post it somewhere, change, etc.)
3. Make contact with book owners
4. To establish cooperation with potential users of the service: foreign language courses, pupils, students, literature lovers
5. Develop a business model (how to promote a site, which books to add initially, in what quantity, how to earn income, distribute it among the project participants)
6. Support financially on pre-agreed terms