← Back to Projects
Fun

Dataset Translator

CLI tool for translating English text datasets into Burmese using the Gemini API with safe batch processing.

↗ GitHub↗ npm
CLINode.jsGemini APInpmNLPBurmese

Key Highlights

  • ◆Batch translation to reduce Gemini API calls
  • ◆Token-aware batching for reliable processing
  • ◆Automatic retry on temporary errors
  • ◆Safe stop + partial save when quota is exceeded
  • ◆Works with CSV datasets (text + label columns)

Overview

A CLI tool for translating English text datasets into Burmese using the Gemini API. Designed for AI/ML dataset preparation with safe batch translation, automatic error handling, and API limit protection. Useful for sentiment analysis, NLP localization, and multilingual training data.

Features

  • –Batch translation to minimize API requests
  • –Token-aware batching for reliable processing
  • –Automatic retry on temporary API errors
  • –Safe stop when quota is exceeded - no data lost
  • –Saves translated and untranslated rows separately
  • –Works with CSV datasets (text + label columns)
  • –Uses Gemini 2.5 Flash for fast translation

Installation & Setup

Install globally

npm install -g dataset-translator

Or run via npx

npx dataset-translator

Run and follow prompts

dataset-translator
# Enter CSV dataset path: example_dataset.csv
# Enter Gemini API Key: **********************

ℹ CSV must have 'text' and 'label' columns. Get a Gemini API key at ai.google.dev

Usage

Expected CSV format and output files:

# Input CSV format:
text,label
I love programming.,2
This is very sad.,0

# Output files (batched):
burmese_dataset_0-457.csv        # Successfully translated rows
burmese_dataset_457-855.csv
untranslated_dataset_855-953.csv # Rows that hit quota limits

Related Projects

FunFun

Burmese Quote Generator

CLI tool to generate emotion-tagged Burmese quotes using the Gemini API in TXT or CSV format.

View project →
FunFun

NEZT CLI

Next & Nuxt EaZy Templates

Scaffold fully configured Next.js and Nuxt.js projects with routing, themes, and pages in minutes.

View project →
AINLP

Elyza Myanmar Chatbot

A rule-based Myanmar language chatbot with emotion detection, served as a Flask REST API.

View project →