I need to convert my website's pages from English to the Urdu language. For this I was using Google's Translation API, but Google translate API is not returning the correct translation of the pages.
What should I to use to get 99% accurate results when translating pages from English Language to Urdu Language?
There are only few parameters that you can specify when using Google Translate API and that can make a difference to your results: source and model parameters:
Source is the language of the source text. If you don't specify it then it will be detected automatically. As your source language is English then I don't think this will be causing any troubles.
Model: As Urdu language is supported by the Neural Machine Translation Model, if you don't specify the model, then nmt model will be used. You can try to use base model, however the nmt one is supposed to "provide improved translation for longer and more complex content".
Maybe expecting the model to get 99% accuracy is expecting it to be almost perfect.
Related
I noticed that transcribing speech in multiple languages with openai whisper speech-to-text library sometimes accurately recognizes inserts in another language and would provide the expected output, for example: 八十多个人 is the same as 八十几个人. So 多 and 几 are interchangeable and they can both mean several.
Yet, the same audio input on a different pass (with the same model, or a smaller/bigger model) would intermittently result in glitches where the entire sentence is being translated rather than transcribed. I.e. a fragment would be translated either into the first or the second language that appears in the audio. With the example input above either the entire sentence would be in English (with Chinese bits translated to English), or the entire sentence would be in Chinese (with the English bits translated to Chinese). Important: in both cases no input language was specified, and no task type was passed (which implies the default --task transcribe).
The docs for whisper mention translation to English as the only available target language (with the option --task translate in the command line version), but there is no mention of translating to other target languages. Yet the behavior mentioned above indicates that the models are capable of doing translation to other languages too.
The question is if there is a known way to configure the models to do just text-to-text translation? Or is the behavior just some sort of glitch that is not something that can be 'exploited' or configured on a lower level that would allow using the models just for text translation between any of the supported languages?
According to a comment in the whisper's issue tracker this might be a possible answer:
From the paper, the dataset that was used did not use any English audio to polish text samples. The dataset was cleaned by using a different model to match spoken language with text language. If they did not match, the sample was excluded. An exception was made for a portion of the training data to match any spoken language to English text (X->en) translation.
So unfortunately there is no direct way, the model wasn't trained on it. For your use case, this can transcribe to English text, but there has to be some an outside system to translate from English text to Polish text.
The --language parameter is defined in the cli as:
--language
language spoken in the audio, specify None
to perform language detection (default: None)
Yet, despite the help text above this can have potentially useful undocumented side effects.
The 'exploit'
The undocumented glitch that was observed is that if you set a source language e.g. es but the audio input contains English then the English part of the input will be translated to Spanish. Parts of the audio input that are not in English will be transcribed although depending on the language it might not always work or it might generate garbage translations.
So the 'exploit' is that the models can be used to parse English audio and then translate it to a supported language.
The behaviour above occurs with the regular transcribe mode (the default, ie. --task transcribe), and is reproducible with both the original whisper implementation in python, as well as the CPU-optimized C++ port whisper.cpp which is using the same models but apparently with different parameters.
The quality of the non-English translation would depend on the language, and seems to be generally of lower quality that translating from English with the open-source huggingface models (e.g. Helsinki-NLP/opus-mt-es-en, facebook/m2m100_418M, facebook/m2m100_1.2B etc).
I am using ngx-translate for Internationalization so how to give file for translation e.g. give English file to translator/third party service for French translation. Considering accuracy can we achieve this please help.
If you have JSON files most translators will have a CAT (computer aided translation) tool which can be used to translate JSON and it will respect the integrity of the JSON format. If the CAT tool does not filter the JSON correctly out of the box there are tweaks which can be made. If you're able to display a sample of one of your files I can provide further feedback. I'm not a translator but I 'prepare' many JSON projects for translators to work on without having to worry about breaking the file.
I read a few papers about machine translation but did not understand them well.
The language models (in Google translate) use phonetics and machine learning as best as I can tell.
My question then becomes is it possible to convert an Arabic word that is phonetically spelled in English to translate the users intended Arabic word?
For instance the word 'Hadith' is an English phonetic of the Arabic word 'حديث'. Can I programmatically go from 'Hadith' to Arabic?
Thanks the Wiki article, there's an entire field of work in the area of Transliteration. There was a Google API for this that was deprecated in 2011 and moved to the Google Input Tools service.
The simplest answer is Buck Walter Transliteration but at first glace a 1:1 mapping doesn't seem like a good enough idea.
I am going to try to see if there's a way to hack the Google Input tools and call it even at CLI level because their online demo works very well
Can someone please give me the official reference to the language (country/region) codes. I'm finding different codes for the same language (es_ES, esp_ESP, etc.) and I can't figure out which one is the right one.
There are several different standards specifying language codes, including ISO-639 with its sub-standards 1-3 and IETF language tags, which describe more of a system of possible codes than the codes themselves.
Which standard is "the right" standard depends on your use case and context. See http://en.wikipedia.org/wiki/Language_codes.
That's because the languages naming coding has different standards, using different number of letters. You might have to chose which standard to use and maybe detect which standard the data source you have is using.
This is a starting point: http://en.wikipedia.org/wiki/List_of_ISO_639-1_codes
These codes are a combination of the specific language as well as the conuntry in which the language is used. So for instance means es_ES spanish_Spain. Another one would be es_AR which would mean spanish_Argentina. For the language code there's the Language Matrix, as for the localisation part you could use the ISO 3166-2 country reference
you can find all the regions code in documentation here
I want to experiment with an idea I have of automatically localizing software, or at least suggesting a reasonable translation if a localized string is not available.
I'm not sure this will be working satisfactorily tomorrow morning but I just wanted to play with this idea.
Does anybody know of a dictionary that is free to use, and is in an easy to parse format, that can help me automatically translate words from English to other European languages (French, German, Spanish, etc)
The FreeDict project has quite a few relatively complete dictionaries. Most are from one language to english or vice versa, but some are between two non-english languages as well.
I don't know any dictionary but would like to point something out. You have to bear in mind that translating is not a direct word to word technique in any sense. The Rules of the language change as well and thus leave sentences unreadable. This is why even companies like Google have trouble making good translation software. Context is very hard to programmatically detect and context means everything in choosing the right word, the right structure and so on.
Maybe use a Translation API, if there is one. Google only seem to do a JavaScript API for Language.
You can't even expect to get a reasonable translation with an automatic method. Translating full texts is too hard for a computer to handle completely correct, translating short phrases correctly is impossible.
Take for example the simple text "Open", without a context it's not even possible to tell if it's a verb or an adjective. I know that at least in german that the verb and the adjective translates into two different words.
Also, computer specific concepts often borrow words from similar concepts outside the computer sphere. Those concepts often have a specific translation, but an automatic translation would sometimes try to translate it as if it was the original meaning, which can give you very strange translations.
After a while of searching i solved the problem by myself start to create my own dictionary. I do a lot of translations in my free time. In the beginning it is really boring work...but after a while you get an really good dicitionary. Some friends of mine using it too...so we all benefit from every new Word we translate.