Overclock.net › Forums › Software, Programming and Coding › Coding and Programming › Translating novel to English using software
New Posts  All Forums:Forum Nav:

Translating novel to English using software

post #1 of 3
Thread Starter 
So my fun project for 2012 is to translate the book Metro 2034 to English.

It was originally written in Russian and an official translation was made to spanish as well as german. Note: this book is available to read for free officially online.

I only know english, but so far I feel my plan is feasible.

Step 1

Run the russian and spanish text through microsofts translator, google translate and babel fish. I have tested software choices such as word magic and babylon and got poor results.

Step 2

Compare the translated texts line by line and determine the best translation for each given sentence through my logical judgement. The multiple sources give me a better idea of what is trying to be communicated. Then select the optimal translation of each sentence to be entered into a master document.

Step 3

Go through the master document and edit for clarity and concision using my trusty sidekick handbook "the elements of style"

Step 4

Upload for consumption by the masses

Anyone have any thoughts on improvement, recommendation of tools to assist or any other suggestions?
Superhans
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Q6600 OC@3.0Ghz, Stock Voltage GIGABYTE GA-G31M-ES2L MSI GTX 950 2GD5T OC 4GB DDR2 800 OEM@833Mhz 
Hard DriveOptical DriveOSMonitor
WD Caviar Black 1TB OEM DVD-R/RW Windows 7 Ultimate x64 Samsung 46" LED 1080P 
KeyboardPowerCaseMouse
Logitech Wave Keyboard corsair 550w Generic Mid-Tower Razer DeathAdder 
  hide details  
Reply
Superhans
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Q6600 OC@3.0Ghz, Stock Voltage GIGABYTE GA-G31M-ES2L MSI GTX 950 2GD5T OC 4GB DDR2 800 OEM@833Mhz 
Hard DriveOptical DriveOSMonitor
WD Caviar Black 1TB OEM DVD-R/RW Windows 7 Ultimate x64 Samsung 46" LED 1080P 
KeyboardPowerCaseMouse
Logitech Wave Keyboard corsair 550w Generic Mid-Tower Razer DeathAdder 
  hide details  
Reply
post #2 of 3
  1. In order to use those services in a program you first would need to get API keys for each service. I know microsoft offers a free SOAP based API, and Google overs a paid for use(1M characters to start) JSON API. To my knowledge yahoo babel fish does not have a API for you to use.
    To pass sentences to each service you would need to make a function that splits apart sentences NOT LINES in order for these translation services to be effective. To do this you would have to set end of sentence delimiters such as .?! . I have no idea how Russian works so I guess you would have to write your own rules to identify a sentence. From there you need to split the result out of the JSON/SOAP responses and store them somewhere to be "compared".
  2. I can't really see how you can automate this part. Since computers are far from competent when it comes to translating things in context you might just be better off reviewing each sentence translated. If you have your heart set on it I would suggest taking a tally of words in one string and comparing them to the other string. From there calculate the percentage of matched words. If say the percentage of same words is more than 80% choose the shorter sentence to write to file.
  3. I can't see a computer being much help.
  4. Depends where you are uploading it too.

Seriously though using translation software to take on a endeavor like this is not the best idea. At the end of the day I can almost guarantee that is going to end up being more trouble than translating the book by hand.
post #3 of 3
As someone who is fluent in both English and Russian, I can't imagine this working. I've translated text before using the above sites and results tend to be far from ideal. Sometimes they are unreadable, others just incorrect. I can almost see this work for non-fiction without advanced technical terms. Novels, however, may use slang, idioms, and other language constructs that translators just don't do well.

Another thing I would point out is the quality of official translations. In general, I would put more faith in an official English translation from a well known publisher than a Spanish one, because my assumption would be that more was invested into making it a quality work. However, having read some official English translations of well-known Russian works (Master and Margarita immediately comes to mind), I would not want to rely on those to translate into a third language.

I will agree that this will far more trouble than translating by hand.
     
CPUMotherboardGraphicsRAM
Intel Core i7-4770K Gigabyte G1.Sniper M5 EVGA GTX 780 SC Crucial Ballistix Sport 2 x 8GB 
Hard DriveHard DriveHard DriveOptical Drive
Samsung 840 Pro 256GB Seagate Barracuda 3TB Seagate Barracuda 4TB Asus BD-ROM 
CoolingOSMonitorKeyboard
Noctua NH-D14 Windows 8 Pro Dell U2713HM Ducky YOTD (MX Brown) 
PowerCaseMouseMouse Pad
SeaSonic SS-760XP2 Fractal Design Arc Mini Roccat Savu SteelSeries 9HD 
Audio
Beyerdynamic DT990 250 Ohm 
CPUMotherboardGraphicsRAM
AMD A6-3500 Asrock A75M HD 6530D Samsung 4GB DDR3 1600  
Hard DriveHard DriveHard DriveOptical Drive
Intel 320 80GB Samsung Spinpoint F4 2TB WD Caviar Green 3TB Sony BD-Rom 
CoolingOSMonitorKeyboard
Noctua NH-C12P SE14 Win 7 Professional 64-bit Samsung LN46C600 Lenovo N5902 
PowerCaseOther
Corsair CX430 Antec Veris Fusion HDHomeRun Prime 
CPUMotherboardGraphicsRAM
AMD Phenom II X3 B55 Asus M5A99X EVO Sapphire HD 7950 G.Skill 8GB DDR3 
Hard DriveOptical DriveCoolingMonitor
Samsung F3 1TB LG DVD-RW CM Hyper 212 Plus Dell U2312HM 
KeyboardPowerCaseMouse
Pink KBT Race (MX Blue) Lepa G700-MA Corsair 500R Razer Abyssus 
Mouse Pad
Razer Goliathus Speed 
  hide details  
Reply
     
CPUMotherboardGraphicsRAM
Intel Core i7-4770K Gigabyte G1.Sniper M5 EVGA GTX 780 SC Crucial Ballistix Sport 2 x 8GB 
Hard DriveHard DriveHard DriveOptical Drive
Samsung 840 Pro 256GB Seagate Barracuda 3TB Seagate Barracuda 4TB Asus BD-ROM 
CoolingOSMonitorKeyboard
Noctua NH-D14 Windows 8 Pro Dell U2713HM Ducky YOTD (MX Brown) 
PowerCaseMouseMouse Pad
SeaSonic SS-760XP2 Fractal Design Arc Mini Roccat Savu SteelSeries 9HD 
Audio
Beyerdynamic DT990 250 Ohm 
CPUMotherboardGraphicsRAM
AMD A6-3500 Asrock A75M HD 6530D Samsung 4GB DDR3 1600  
Hard DriveHard DriveHard DriveOptical Drive
Intel 320 80GB Samsung Spinpoint F4 2TB WD Caviar Green 3TB Sony BD-Rom 
CoolingOSMonitorKeyboard
Noctua NH-C12P SE14 Win 7 Professional 64-bit Samsung LN46C600 Lenovo N5902 
PowerCaseOther
Corsair CX430 Antec Veris Fusion HDHomeRun Prime 
CPUMotherboardGraphicsRAM
AMD Phenom II X3 B55 Asus M5A99X EVO Sapphire HD 7950 G.Skill 8GB DDR3 
Hard DriveOptical DriveCoolingMonitor
Samsung F3 1TB LG DVD-RW CM Hyper 212 Plus Dell U2312HM 
KeyboardPowerCaseMouse
Pink KBT Race (MX Blue) Lepa G700-MA Corsair 500R Razer Abyssus 
Mouse Pad
Razer Goliathus Speed 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Coding and Programming
Overclock.net › Forums › Software, Programming and Coding › Coding and Programming › Translating novel to English using software