API Guides & TutorialsText Processing

Text Translation API

View as Markdown

Translating a file (PDF, Word, Excel, PowerPoint, or HTML) with layout preserved? Use the Document Translation API instead of sending file text through this endpoint.

Document Translation

Upload once, translate into up to 12 languages per job, and export in the same format as the source (PDF→PDF, DOCX→DOCX).

Translate a PDF

End-to-end walkthrough: create a job, upload your file, poll progress, and download the translated document.

Translation Types

English to Indic

Translate from English to various Indian languages with support for different translation modes.

Indic to English

Convert Indian languages to English with high accuracy and natural output.

Indic to Indic

Translate between different Indian languages while preserving context and meaning.

Translation Modes

Formal

Highly professional, uses pure language forms. Ideal for official documents and legal papers.

Classic-Colloquial

Balanced mix of languages, slightly informal. Perfect for business emails and general communication.

Modern-Colloquial

Casual and direct style with mixed language. Best for chatbots and social media content.

Code-Mixed

Mixes English and the target language within the same sentence, mirroring how people actually speak. Available with mayura:v1 via mode="code-mixed".

Code Examples

from sarvamai import SarvamAI
client = SarvamAI(
    api_subscription_key="YOUR_SARVAM_API_KEY"
)
response = client.text.translate(
    input="Hello, how are you?",
    source_language_code="en-IN",
    target_language_code="hi-IN",
    speaker_gender="Male"
)
print(response)

API Features

Translation Options
  • Multiple Indian languages support
  • Four translation modes
  • Gender-specific translations
  • Code-mixed text support
Output Formats
  • Multiple script options
  • Native/International numerals
  • Customizable formatting
  • Transliteration support
Advanced Features
  • Automatic language detection
  • Context preservation
  • Entity handling

API Response Format

FieldTypeDescription
request_idstringUnique identifier for the request
translated_textstringTranslated text in the target language
source_language_codestringDetected or provided source language (BCP-47 format)

Supported languages:

  • mayura:v1: bn-IN, en-IN, gu-IN, hi-IN, kn-IN, ml-IN, mr-IN, od-IN, pa-IN, ta-IN, te-IN
  • sarvam-translate:v1: All mayura languages + as-IN, brx-IN, doi-IN, kok-IN, ks-IN, mai-IN, mni-IN, ne-IN, sa-IN, sat-IN, sd-IN, ur-IN
{
  "request_id": "20241115_12345678-1234-5678-1234-567812345678",
  "translated_text": "नमस्ते, आप कैसे हैं?",
  "source_language_code": "en-IN"
}

With Roman Script Output:

{
  "request_id": "20241115_12345678-1234-5678-1234-567812345678",
  "translated_text": "namaste, aap kaise hain?",
  "source_language_code": "en-IN"
}

With Native Numerals:

{
  "request_id": "20241115_12345678-1234-5678-1234-567812345678",
  "translated_text": "मेरा फोन नंबर है ९८४०९५०९५०",
  "source_language_code": "en-IN"
}

Auto-Detected Source Language:

{
  "request_id": "20241115_12345678-1234-5678-1234-567812345678",
  "translated_text": "Hello, how are you?",
  "source_language_code": "hi-IN"
}

Error Responses

All errors return a JSON object with an error field (message, code, request_id). The full error-code table, retry guidance, and SDK exception reference live on the central Errors & Troubleshooting page.

Errors specific to this endpoint:

HTTP StatusError CodeWhen This HappensWhat To Do
400invalid_request_errorMissing required parameters or malformed requestCheck input, source_language_code, target_language_code
422unprocessable_entity_errorText too long or unsupported language pairKeep text under 1000 chars (mayura:v1) or 2000 chars (sarvam-translate:v1)
from sarvamai import SarvamAI
from sarvamai.core.api_error import ApiError
client = SarvamAI(api_subscription_key="YOUR_SARVAM_API_KEY")
try:
    response = client.text.translate(
        input="Hello, how are you?",
        source_language_code="en-IN",
        target_language_code="hi-IN"
    )
    print(response.translated_text)
except ApiError as e:
    if e.status_code == 400:
        print(f"Bad request: {e.body}")
    elif e.status_code == 403:
        print("Invalid API key. Check your credentials.")
    elif e.status_code == 422:
        print(f"Invalid parameters: {e.body}")
    elif e.status_code == 429:
        print("Rate limit exceeded. Wait and retry.")
    else:
        print(f"Error {e.status_code}: {e.body}")

Check out our detailed API Reference to explore Translation and all available options.

Need help with translation? Contact us on discord for guidance.