Transliteration: Get started in 8 steps on the NeuralSpace Platform
Introduction
For languages that don't use the Latin (English) alphabet, e.g., Arabic, Hindi, Punjabi, Sinhala, typing can be challenging as keyboards/keypads often default to Latin characters. That makes creating content in these vernacular languages difficult. With Transliteration, you can create content in these languages using your Latin keypad. It transforms a word from one alphabet to the other phonetically.
For instance, you type a word on the Latin keypad the way you would pronounce it in Punjabi, then using transliteration you can convert that into the Punjabi alphabet.
Read more about NeuralSpace’s Transliteration service: features, use-cases and language support here.
In this short tutorial, we will explain to you how to get started with Transliteration, on the NeuralSpace Platform.
It’s an easy 8-step process and you will have trained your first Transliteration model with AutoNLP in one of NeuralSpace’s 87 supported languages. Best of all, we will use NeuralSpace’s no-code web interface and achieve state-of-the-art results without writing a single line of code.
Let's Get Started
1. SIGN-UP
Without any credit card or other payment details, sign up to NeuralSpace here, activate your account through your verification email and log in.
2. TAKE THE TOUR
After logging in, you will be welcomed by a virtual tour through which you can Import a dataset, then Train and Deploy a model which you can test.
3. SELECT TRANSLITERATION
Select Transliteration from All Services in the column menu on the left. Then click on Projects.
4. CREATE PROJECT
Navigate to the top right corner and click on Create project. If you want to read more about the concepts behind projects, check out our Docs.
After you click on Create Project, you will see three options as can be seen in the image below. For this tutorial, we will create a project using the "Import from NeuralSpace Datasets" option.
5. IMPORT DATASET
To import a dataset, select a Source and Target Language you want to train your model in.
Here we select English as the source and Hindi as the target language. Next, select a dataset from the available options, click on Import and give your project a name.
Now, you will be redirected to the Projects page where you can see all your current Translietartion projects. Wait for a couple of minutes as your dataset gets uploaded. Click on the project tile once the dataset is 100% uploaded.
6. TRAIN
Now, you can see your project details. You are now prepared to train your custom model with AutoNLP. Select the number of training jobs (read more about what training jobs are here), click on Train with AutoNLP and you are ready to — CLICK, TRAIN, CHILL at its best.
You can follow the progress in the Model List at the bottom of your Project Details page.
The Training Status will go from Queued to Training to Completed within a few seconds, but you may need to wait a bit longer for larger datasets. Whereas a small dataset with 20 examples only takes about 2 minutes to train, a dataset with 5000 examples will be trained in about 15 minutes.
7. DEPLOY
Once your model is trained and you can deploy it now by clicking on the Deploy with AutoMLOps button.
8. TEST
Once it’s deployed, you can evaluate it by clicking on the model in the Model List and then Test model in the right menu. Type in any sentence you want to test your model on and see its output. As you write a word you will get an option to choose from various transliterated suggestions from a dropdown.
That was it! 8 simple steps to train and deploy a state-of-the-art transformer-based deep learning model and evaluate it in code-mixed Hinglish (English + Hindi).
-------
Get started now: NeuralSpace Platform
Read about other tutorials in our Docs and sign up to the NeuralSpace Community to get involved and collaborate with fellow users.
Happy NLP!