BhashiniClient

A Python client library for interacting with Bhashini services, including Automatic Speech Recognition (ASR), Neural Machine Translation (NMT), and Text-to-Speech (TTS) for 13 indian languages.

• Test the API online here and create a free API key and USER ID from profile.
• For Detailed API DOCS click here

Table of Contents

• Features
• Requirements
• Installation
• Usage
  • Initialization
  • Examples
    • ASR Example
    • Translation Example
    • TTS Example
  • Methods
    • list_available_languages(task_type)
    • get_supported_voices(source_language)
    • asr(audio_content, source_language, audio_format='wav', sampling_rate=16000)
    • translate(text, source_language, target_language)
    • tts(text, source_language, gender='female', sampling_rate=8000)
• License

Features

• ASR (Automatic Speech Recognition): Convert speech audio to text.
  • Supported Formats: .wav, .mp3, .flac, .ogg
  • Sampling Rates: 8000 Hz, 16000 Hz, 48000 Hz
• Translation: Translate text between supported language pairs.
• TTS (Text-to-Speech): Convert text to speech audio with gender options.

Requirements

• Python 3.x
• requests library

Install the required Python package:

pip install requests

Installation

Copy the BhashiniClient class into your project or save it as a module (e.g., bhashini_client.py).

Usage

Initialization

Initialize the BhashiniClient with your user credentials and pipeline ID.

from bhashini_client import BhashiniClient

USER_ID = 'your_user_id'
API_KEY = 'your_ulca_api_key'
PIPELINE_ID = 'your_pipeline_id'  # Optional, defaults to predefined PIPELINE_ID

client = BhashiniClient(USER_ID, API_KEY, PIPELINE_ID)

Get free api key and user id from here

Examples

ASR Example

# Initialize the client
client = BhashiniClient(USER_ID, API_KEY)

# Perform ASR
with open('speech.wav', 'rb') as f:
    audio_content = f.read()

asr_result = client.asr(
    audio_content,
    source_language='hi',
    audio_format='wav',       # Supported formats: 'wav', 'mp3', 'flac', 'ogg'
    sampling_rate=16000       # Supported rates: 8000, 16000, 48000
)
print("Transcribed Text:", asr_result['pipelineResponse'][0]['output'][0]['source'])

Translation Example

# Initialize the client
client = BhashiniClient(USER_ID, API_KEY)

# Translate text
text = 'मेरा नाम विहिर है।'
translation_result = client.translate(text, source_language='hi', target_language='en')
print("Translated Text:", translation_result['pipelineResponse'][0]['output'][0]['target'])

TTS Example

# Initialize the client
client = BhashiniClient(USER_ID, API_KEY)

# Convert text to speech
text = 'नमस्ते दुनिया'
tts_result = client.tts(
    text,
    source_language='hi',
    gender='female',
    sampling_rate=8000       # Supported rates: 8000, 16000, 48000
)

# Save the audio file
audio_base64 = tts_result['pipelineResponse'][0]['audio'][0]['audioContent']
audio_data = base64.b64decode(audio_base64)
with open('output.wav', 'wb') as f:
    f.write(audio_data)

Methods

list_available_languages(task_type)

Lists the available languages for a specified task.

• Parameters:

  • task_type (str): The task type ('asr', 'translation', or 'tts').

• Returns:

  • list or dict: A list of languages or a dictionary of language pairs for translation.

Usage Example:

# FOR TTS
tts_languages = client.list_available_languages('tts')
print("TTS Languages:", tts_languages)

# For ASR
asr_languages = client.list_available_languages('asr')
print("ASR Languages:", asr_languages)

# For Translation
translation_languages = client.list_available_languages('translation')
print("Translation Languages:", translation_languages)

TTS Languages: ['en', 'as', 'brx', 'gu', 'hi', 'kn', 'ml', 'mni', 'mr', 'or', 'pa', 'ta', 'te', 'bn']
ASR Languages: ['bn', 'en', 'gu', 'hi', 'kn', 'ml', 'mr', 'or', 'pa', 'sa', 'ta', 'te', 'ur']
Translation Languages: {'bn': ['en', 'as', 'brx', 'gu', 'hi', 'kn', 'ml', 'mni', 'mr', 'or', 'pa', 'ta', 'te'], 'en': ['as', 'bn', 'brx', 'gu', 'hi', 'kn', 'ml', 'mni', 'mr', 'or', 'pa', 'ta', 'te'], 'gu': ['en', 'as', 'bn', 'brx', 'hi', 'kn', 'ml', 'mni', 'mr', 'or', 'pa', 'ta', 'te'], 'hi': ['en', 'as', 'bn', 'brx', 'gu', 'kn', 'ml', 'mni', 'mr', 'or', 'pa', 'ta', 'te'], 'kn': ['en', 'as', 'bn', 'brx', 'gu', 'hi', 'ml', 'mni', 'mr', 'or', 'pa', 'ta', 'te'], 'ml': ['en', 'as', 'bn', 'brx', 'gu', 'hi', 'kn', 'mni', 'mr', 'or', 'pa', 'ta', 'te'], 'mr': ['en', 'as', 'bn', 'brx', 'gu', 'hi', 'kn', 'ml', 'mni', 'or', 'pa', 'ta', 'te'], 'or': ['en', 'as', 'bn', 'brx', 'gu', 'hi', 'kn', 'ml', 'mni', 'mr', 'pa', 'ta', 'te'], 'pa': ['en', 'as', 'bn', 'brx', 'gu', 'hi', 'kn', 'ml', 'mni', 'mr', 'or', 'ta', 'te'], 'sa': ['en', 'as', 'bn', 'brx', 'gu', 'hi', 'kn', 'ml', 'mni', 'mr', 'or', 'pa', 'ta', 'te'], 'ta': ['en', 'as', 'bn', 'brx', 'gu', 'hi', 'kn', 'ml', 'mni', 'mr', 'or', 'pa', 'te'], 'te': ['en', 'as', 'bn', 'brx', 'gu', 'hi', 'kn', 'ml', 'mni', 'mr', 'or', 'pa', 'ta'], 'ur': ['en', 'as', 'bn', 'brx', 'gu', 'hi', 'kn', 'ml', 'mni', 'mr', 'or', 'pa', 'ta', 'te']}

get_supported_voices(source_language)

Retrieves supported voices for TTS in the specified language.

• Parameters:

  • source_language (str): Language code (e.g., 'hi' for Hindi).

• Returns:

  • list: Supported voices (['male', 'female']).

Usage Example:

voices = client.get_supported_voices('hi')
print("Supported Voices for Hindi TTS:", voices)

asr(audio_content, source_language, audio_format='wav', sampling_rate=16000)

Performs ASR on the provided audio content.

• Parameters:

  • audio_content (bytes): Audio content in bytes.
  • source_language (str): Language code of the audio.
  • audio_format (str): Audio format ('wav', 'mp3', 'flac', 'ogg').
  • sampling_rate (int): Sampling rate in Hz (8000, 16000, 48000).

• Returns:

  • dict: ASR response from the API.

Usage Example:

with open('audio.wav', 'rb') as f:
    audio_content = f.read()

asr_result = client.asr(
    audio_content,
    source_language='hi',
    audio_format='wav',       # Supported formats: 'wav', 'mp3', 'flac', 'ogg'
    sampling_rate=16000       # Supported rates: 8000, 16000, 48000
)
print("ASR Result:", asr_result)

translate(text, source_language, target_language)

Translates text from the source language to the target language.

• Parameters:

  • text (str): Text to translate.
  • source_language (str): Source language code.
  • target_language (str): Target language code.

• Returns:

  • dict: Translation response from the API.

Usage Example:

translation_result = client.translate(
    'मेरा नाम विहिर है।',
    source_language='hi',
    target_language='en'
)
print("Translation Result:", translation_result)

tts(text, source_language, gender='female', sampling_rate=8000)

Converts text to speech in the specified language and gender.

• Parameters:

  • text (str): Text to convert.
  • source_language (str): Language code.
  • gender (str): 'male' or 'female'.
  • sampling_rate (int): Sampling rate in Hz (8000, 16000, 48000).

• Returns:

  • dict: TTS response from the API.

Usage Example:

tts_result = client.tts(
    'હેલો વર્લ્ડ',
    source_language='gu',
    gender='female',
    sampling_rate=16000       # Supported rates: 8000, 16000, 48000
)

# Save the audio output
audio_base64 = tts_result['pipelineResponse'][0]['audio'][0]['audioContent']
audio_data = base64.b64decode(audio_base64)
with open('output_audio.wav', 'wb') as f:
    f.write(audio_data)

License

This project is licensed under the MIT License.

---

Feel free to contribute to this project by submitting issues or pull requests.