Zaban Python Client

Official Python client for Zaban API - providing access to Zaban services for Indian languages including translation, text-to-speech, speech-to-text, and transliteration.

🌟 Features

• Translation: Powered by IndicTrans2, supporting 22 Indian languages
• Text-to-Speech (TTS): Convert text to natural-sounding speech
• Speech-to-Text (STT): Transcribe audio to text
• Transliteration: Convert text between different scripts
• Async Support: Full async/await support for concurrent operations
• Type Safe: Fully typed with Pydantic models
• Simple API: Clean, intuitive interface similar to OpenAI's client

📦 Installation

pip install zaban

🚀 Quick Start

from zaban import Zaban

# Initialize client with API key
client = Zaban(api_key="sk-your-api-key")

# Or use environment variable ZABAN_API_KEY
client = Zaban()

# Translate text
result = client.translation.create(
    text="Hello, how are you?",
    source_lang="eng_Latn",
    target_lang="hin_Deva"
)
print(result.translated_text)  # "आप कैसे हैं?"

📖 Usage Examples

Translation

# English to Hindi
result = client.translation.create(
    text="Hello, how are you?",
    source_lang="eng_Latn",
    target_lang="hin_Deva"
)
print(result.translated_text)  # "आप कैसे हैं?"

# Auto-detect source language
result = client.translation.create(
    text="Hello",
    target_lang="hin_Deva",
    auto_detect=True
)

# Using convenience method
result = client.translation.translate(
    "Hello",
    to="hin_Deva",
    from_="eng_Latn"
)

Text-to-Speech (TTS)

# Generate speech from text
audio = client.audio.speech.create(
    text="नमस्ते दुनिया",
    lang="hi",
    speaker="female",
    format="wav"
)

# Save to file
audio.save("output.wav")

# Or get bytes directly
audio_bytes = audio.content

Speech-to-Text (STT)

# Transcribe from file path
transcription = client.audio.transcriptions.create(
    audio="audio.wav",
    lang="hi"
)
print(transcription.text)

# From file object
with open("audio.wav", "rb") as f:
    transcription = client.audio.transcriptions.create(
        audio=f,
        lang="hi"
    )

# From audio URL
transcription = client.audio.transcriptions.create(
    audio_url="https://example.com/audio.wav",
    lang="hi"
)

Transliteration

# Transliterate from Latin to Devanagari
result = client.transliteration.create(
    text="namaste",
    source_script="latn",
    target_script="deva",
    lang="hi",
    topk=3
)
print(result.top)  # "नमस्ते"
print(result.results)  # ["नमस्ते", "नमस्ते", "नमस्ते"]

⚡ Async Support

import asyncio
from zaban import AsyncZaban

async def main():
    # Initialize async client
    async with AsyncZaban(api_key="sk-your-api-key") as client:
        # Single translation
        result = await client.translation.create(
            text="Hello",
            target_lang="hin_Deva",
            auto_detect=True
        )
        print(result.translated_text)
        
        # Batch translations (concurrent)
        texts = ["Hello", "Goodbye", "Thank you"]
        tasks = [
            client.translation.create(text=t, target_lang="hin_Deva", auto_detect=True)
            for t in texts
        ]
        results = await asyncio.gather(*tasks)
        
        for result in results:
            print(result.translated_text)

asyncio.run(main())

🌐 Supported Languages

The Zaban API supports 22 Indian languages plus English using BCP-47 format with script:

Language	Code	Script
English	eng_Latn	Latin
Hindi	hin_Deva	Devanagari
Bengali	ben_Beng	Bengali
Tamil	tam_Taml	Tamil
Telugu	tel_Telu	Telugu
Gujarati	guj_Gujr	Gujarati
Kannada	kan_Knda	Kannada
Malayalam	mal_Mlym	Malayalam
Marathi	mar_Deva	Devanagari
Punjabi	pan_Guru	Gurmukhi
Oriya	ory_Orya	Oriya
Assamese	asm_Beng	Bengali
Urdu	urd_Arab	Arabic

[See full list in documentation]

🔧 Configuration

Environment Variables

# Set API key
export ZABAN_API_KEY="sk-your-api-key"

# Alternative
export ZABAN_KEY="sk-your-api-key"

Client Options

client = Zaban(
    api_key="sk-your-api-key",       # API key
    base_url="http://localhost:8000/api/v1",  # API base URL
    timeout=30.0,                     # Request timeout in seconds
    max_retries=2,                    # Max retries for failed requests
)

🛡️ Error Handling

from zaban import (
    Zaban,
    ZabanError,
    AuthenticationError,
    RateLimitError,
    ValidationError,
    APIError,
    TimeoutError,
    ConnectionError,
)

try:
    result = client.translation.create(
        text="Hello",
        target_lang="hin_Deva"
    )
except AuthenticationError as e:
    print(f"Invalid API key: {e}")
except RateLimitError as e:
    print(f"Rate limit exceeded: {e}")
except ValidationError as e:
    print(f"Invalid request: {e}")
except TimeoutError as e:
    print(f"Request timed out: {e}")
except ConnectionError as e:
    print(f"Connection failed: {e}")
except ZabanError as e:
    print(f"API error: {e}")

🔑 Getting an API Key

1. Sign up at [Zaban Dashboard]
2. Navigate to API Keys section
3. Create a new API key
4. Copy the key (starts with sk-)
5. Use it in your client initialization

📚 Examples

Check out the examples/ directory for more usage examples:

• basic_usage.py - Basic usage examples
• async_usage.py - Async/await examples
• batch_translation.py - Batch processing

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🔗 Links

• GitHub Repository
• Zaban API Documentation
• AI4Bharat
• IndicTrans2

💬 Support

• GitHub Issues: Report bugs or request features
• Email: support@zaban.ai

🙏 Acknowledgments

Built with ❤️ using AI4Bharat's amazing open-source models:

• IndicTrans2 for translation
• AI4Bharat TTS/STT services for speech processing

---

Note: This is an unofficial client. For official API documentation.