Spaces:
Running
Running
File size: 16,700 Bytes
7e87316 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 |
import requests
import json
import base64
class BhashiniClient:
"""
A client for interacting with Bhashini's ASR, NMT, and TTS services.
Methods:
list_available_languages(task_type): Lists available languages for a given task.
get_supported_voices(source_language): Gets supported genders for TTS in a language.
asr(audio_content, source_language, audio_format='wav', sampling_rate=16000): Performs ASR.
translate(text, source_language, target_language): Translates text from source to target language.
tts(text, source_language, gender='female', sampling_rate=8000): Performs TTS.
"""
PIPELINE_CONFIG_ENDPOINT = "https://meity-auth.ulcacontrib.org/ulca/apis/v0/model/getModelsPipeline"
INFERENCE_ENDPOINT = "https://dhruva-api.bhashini.gov.in/services/inference/pipeline"
PIPELINE_ID = "64392f96daac500b55c543cd"
def __init__(self, user_id, api_key, pipeline_id = PIPELINE_ID):
"""
Initializes the BhashiniClient with user credentials and pipeline ID.
Args:
user_id (str): Your user ID.
api_key (str): Your ULCA API key.
pipeline_id (str): The pipeline ID.
Raises:
Exception: If the pipeline configuration retrieval fails.
"""
self.user_id = user_id
self.api_key = api_key
self.pipeline_id = pipeline_id
self.headers = {
"Content-Type": "application/json",
"userID": self.user_id,
"ulcaApiKey": self.api_key
}
self.config = self._get_pipeline_config()
self.pipeline_data = self._parse_pipeline_config(self.config)
self.inference_api_key = self.pipeline_data['inferenceApiKey']
def _get_pipeline_config(self):
"""
Retrieves the pipeline configuration.
Returns:
dict: The pipeline configuration.
Raises:
Exception: If the request fails.
"""
payload = {
"pipelineTasks": [
{"taskType": "asr"},
{"taskType": "translation"},
{"taskType": "tts"}
],
"pipelineRequestConfig": {
"pipelineId": self.pipeline_id
}
}
response = requests.post(
self.PIPELINE_CONFIG_ENDPOINT,
headers=self.headers,
data=json.dumps(payload)
)
response.raise_for_status()
return response.json()
def _parse_pipeline_config(self, config):
"""
Parses the pipeline configuration and extracts necessary information.
Args:
config (dict): The pipeline configuration.
Returns:
dict: Parsed pipeline data.
"""
inference_api_key = config['pipelineInferenceAPIEndPoint']['inferenceApiKey']['value']
callback_url = config['pipelineInferenceAPIEndPoint']['callbackUrl']
pipeline_data = {
'asr': {},
'tts': {},
'translation': {},
'inferenceApiKey': inference_api_key,
'callbackUrl': callback_url
}
for pipeline in config['pipelineResponseConfig']:
task_type = pipeline['taskType']
if task_type in ['asr', 'translation', 'tts']:
for language_config in pipeline['config']:
source_language = language_config['language']['sourceLanguage']
if task_type != 'translation':
if source_language not in pipeline_data[task_type]:
pipeline_data[task_type][source_language] = []
language_info = {
'serviceId': language_config['serviceId'],
'sourceScriptCode': language_config['language'].get('sourceScriptCode')
}
if task_type == 'tts':
language_info['supportedVoices'] = language_config.get('supportedVoices', [])
pipeline_data[task_type][source_language].append(language_info)
else:
target_language = language_config['language']['targetLanguage']
if source_language not in pipeline_data[task_type]:
pipeline_data[task_type][source_language] = {}
if target_language not in pipeline_data[task_type][source_language]:
pipeline_data[task_type][source_language][target_language] = []
language_info = {
'serviceId': language_config['serviceId'],
'sourceScriptCode': language_config['language'].get('sourceScriptCode'),
'targetScriptCode': language_config['language'].get('targetScriptCode')
}
pipeline_data[task_type][source_language][target_language].append(language_info)
return pipeline_data
def list_available_languages(self, task_type):
"""
Lists the available languages for the specified task.
Args:
task_type (str): The task type ('asr', 'translation', or 'tts').
Returns:
list or dict: A list of available languages, or a dictionary for translation.
Raises:
ValueError: If an invalid task type is provided.
Usage Example:
client = BhashiniClient(user_id, api_key, pipeline_id)
asr_languages = client.list_available_languages('asr')
print("Available ASR Languages:", asr_languages)
translation_languages = client.list_available_languages('translation')
print("Available Translation Languages:", translation_languages)
"""
if task_type not in ['asr', 'translation', 'tts']:
raise ValueError("Invalid task type. Choose from 'asr', 'translation', or 'tts'.")
if task_type == 'translation':
languages = {}
for src_lang in self.pipeline_data['translation']:
languages[src_lang] = list(self.pipeline_data['translation'][src_lang].keys())
return languages
else:
return list(self.pipeline_data[task_type].keys())
def get_supported_voices(self, source_language):
"""
Returns the supported genders for TTS in the specified language.
Args:
source_language (str): The language code (e.g., 'hi' for Hindi).
Returns:
list: A list of supported genders (e.g., ['male', 'female']).
Raises:
ValueError: If TTS is not supported for the language.
Usage Example:
client = BhashiniClient(user_id, api_key, pipeline_id)
voices = client.get_supported_voices('hi')
print("Supported voices for Hindi TTS:", voices)
"""
if source_language not in self.pipeline_data['tts']:
available_languages = ', '.join(self.list_available_languages('tts'))
raise ValueError(
f"TTS not supported for language '{source_language}'. "
f"Available languages: {available_languages}"
)
service_info = self.pipeline_data['tts'][source_language][0]
supported_voices = service_info.get('supportedVoices', [])
return supported_voices
def asr(self, audio_content, source_language, audio_format='wav', sampling_rate=16000):
"""
Performs Automatic Speech Recognition on the provided audio content.
Args:
audio_content (bytes): The audio content in bytes.
source_language (str): The language code of the audio (e.g., 'hi' for Hindi).
audio_format (str): supported formats of audio content: ('wav', 'mp3', 'flac', 'ogg'.)
sampling_rate (int): The sampling rate of the audio in Hz.
Returns:
dict: The ASR response from the API.
Raises:
ValueError: If the language is not supported.
Exception: If the API request fails.
Usage Example:
client = BhashiniClient(user_id, api_key, pipeline_id)
with open('audio.wav', 'rb') as f:
audio_content = f.read()
asr_result = client.asr(audio_content, source_language='hi', audio_format='wav')
print("ASR Result:", asr_result)
"""
if source_language not in self.pipeline_data['asr']:
available_languages = ', '.join(self.list_available_languages('asr'))
raise ValueError(
f"ASR not supported for language '{source_language}'. "
f"Available languages: {available_languages}"
)
service_info = self.pipeline_data['asr'][source_language][0]
service_id = service_info['serviceId']
payload = {
"pipelineTasks": [
{
"taskType": "asr",
"config": {
"language": {
"sourceLanguage": source_language
},
"serviceId": service_id,
"audioFormat": audio_format,
"samplingRate": sampling_rate
}
}
],
"inputData": {
"audio": [
{
"audioContent": base64.b64encode(audio_content).decode('utf-8')
}
]
}
}
headers = {
'Accept': '*/*',
'Authorization': self.inference_api_key,
'Content-Type': 'application/json'
}
response = requests.post(
self.INFERENCE_ENDPOINT,
headers=headers,
data=json.dumps(payload)
)
self._handle_response_errors(response)
return response.json()
def translate(self, text, source_language, target_language):
"""
Translates the provided text from the source language to the target language.
Args:
text (str): The text to translate.
source_language (str): The source language code.
target_language (str): The target language code.
Returns:
dict: The translation response from the API.
Raises:
ValueError: If the language pair is not supported.
Exception: If the API request fails.
Usage Example:
client = BhashiniClient(user_id, api_key, pipeline_id)
translation_result = client.translate(
'मेरा नाम विहिर है।',
source_language='hi',
target_language='gu'
)
print("Translation Result:", translation_result)
"""
if source_language not in self.pipeline_data['translation']:
available_languages = ', '.join(self.list_available_languages('translation').keys())
raise ValueError(
f"Translation not supported from language '{source_language}'. "
f"Available source languages: {available_languages}"
)
if target_language not in self.pipeline_data['translation'][source_language]:
available_targets = ', '.join(self.pipeline_data['translation'][source_language].keys())
raise ValueError(
f"Translation from '{source_language}' to '{target_language}' not supported. "
f"Available target languages for '{source_language}': {available_targets}"
)
service_info = self.pipeline_data['translation'][source_language][target_language][0]
service_id = service_info['serviceId']
payload = {
"pipelineTasks": [
{
"taskType": "translation",
"config": {
"language": {
"sourceLanguage": source_language,
"targetLanguage": target_language
},
"serviceId": service_id
}
}
],
"inputData": {
"input": [
{
"source": text
}
]
}
}
headers = {
'Accept': '*/*',
'Authorization': self.inference_api_key,
'Content-Type': 'application/json'
}
response = requests.post(
self.INFERENCE_ENDPOINT,
headers=headers,
data=json.dumps(payload)
)
self._handle_response_errors(response)
return response.json()
def tts(self, text, source_language, gender='female', sampling_rate=8000):
"""
Converts the provided text to speech in the specified language.
Args:
text (str): The text to convert to speech.
source_language (str): The language code of the text.
gender (str): The desired voice gender ('male' or 'female').
sampling_rate (int): The sampling rate in Hz.
Returns:
dict: The TTS response from the API.
Raises:
ValueError: If the language or gender is not supported.
Exception: If the API request fails.
Usage Example:
client = BhashiniClient(user_id, api_key, pipeline_id)
tts_result = client.tts(
'હેલો વર્લ્ડ',
source_language='gu',
gender='female'
)
# 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)
"""
if source_language not in self.pipeline_data['tts']:
available_languages = ', '.join(self.list_available_languages('tts'))
raise ValueError(
f"TTS not supported for language '{source_language}'. "
f"Available languages: {available_languages}"
)
service_info = self.pipeline_data['tts'][source_language][0]
service_id = service_info['serviceId']
supported_voices = service_info.get('supportedVoices', [])
if gender not in ['male', 'female']:
raise ValueError("Gender must be 'male' or 'female'.")
if supported_voices and gender not in supported_voices:
available_genders = ', '.join(supported_voices)
raise ValueError(
f"Gender '{gender}' not supported for language '{source_language}'. "
f"Available genders: {available_genders}"
)
payload = {
"pipelineTasks": [
{
"taskType": "tts",
"config": {
"language": {
"sourceLanguage": source_language
},
"serviceId": service_id,
"gender": gender,
"samplingRate": sampling_rate
}
}
],
"inputData": {
"input": [
{
"source": text
}
]
}
}
headers = {
'Accept': '*/*',
'Authorization': self.inference_api_key,
'Content-Type': 'application/json'
}
response = requests.post(
self.INFERENCE_ENDPOINT,
headers=headers,
data=json.dumps(payload)
)
self._handle_response_errors(response)
return response.json()
def _handle_response_errors(self, response):
"""
Handles errors in the response.
Args:
response (requests.Response): The response object.
Raises:
Exception: If an HTTP error occurs.
"""
try:
response.raise_for_status()
except requests.HTTPError as http_err:
try:
error_info = response.json()
error_message = error_info.get('message', 'An error occurred.')
except json.JSONDecodeError:
error_message = response.text
raise Exception(f"HTTP error occurred: {error_message}") from http_err
|