dashscope 1.20.8__py3-none-any.whl → 1.20.10__py3-none-any.whl

dashscope/aigc/image_synthesis.py

@@ -6,6 +6,7 @@ from dashscope.client.base_api import BaseAsyncApi
 from dashscope.common.constants import IMAGES, NEGATIVE_PROMPT, PROMPT
 from dashscope.common.error import InputRequired
 from dashscope.common.utils import _get_task_group_and_task
+from dashscope.utils.oss_utils import check_and_upload_local
 
 
 class ImageSynthesis(BaseAsyncApi):
@@ -38,7 +39,8 @@ class ImageSynthesis(BaseAsyncApi):
             images (List[str]): The input list of images url,
                 currently not supported.
             api_key (str, optional): The api api_key. Defaults to None.
-            sketch_image_url (str, optional): Only for wanx-sketch-to-image-v1.
+            sketch_image_url (str, optional): Only for wanx-sketch-to-image-v1,
+                can be local file.
                 Defaults to None.
             workspace (str): The dashscope workspace id.
             extra_input (Dict): The extra input parameters.
@@ -120,17 +122,31 @@ class ImageSynthesis(BaseAsyncApi):
             raise InputRequired('prompt is required!')
         task_group, function = _get_task_group_and_task(__name__)
         input = {PROMPT: prompt}
+        has_upload = False
         if negative_prompt is not None:
             input[NEGATIVE_PROMPT] = negative_prompt
         if images is not None:
             input[IMAGES] = images
         if sketch_image_url is not None and sketch_image_url:
+            is_upload, sketch_image_url = check_and_upload_local(
+                model, sketch_image_url, api_key)
+            if is_upload:
+                has_upload = True
             input['sketch_image_url'] = sketch_image_url
         if ref_img is not None and ref_img:
+            is_upload, ref_img = check_and_upload_local(
+                model, ref_img, api_key)
+            if is_upload:
+                has_upload = True
             input['ref_img'] = ref_img
         if extra_input is not None and extra_input:
             input = {**input, **extra_input}
 
+        if has_upload:
+            headers = kwargs.pop('headers', {})
+            headers['X-DashScope-OssResourceResolve'] = 'enable'
+            kwargs['headers'] = headers
+
         response = super().async_call(
             model=model,
             task_group=task_group,

dashscope/audio/asr/__init__.py

@@ -1,8 +1,9 @@
 from .asr_phrase_manager import AsrPhraseManager
 from .recognition import Recognition, RecognitionCallback, RecognitionResult
 from .transcription import Transcription
+from .vocabulary import VocabularyService, VocabularyServiceException
 
 __all__ = [
-    Transcription, Recognition, RecognitionCallback, RecognitionResult,
-    AsrPhraseManager
+    'Transcription', 'Recognition', 'RecognitionCallback', 'RecognitionResult',
+    'AsrPhraseManager', 'VocabularyServiceException', 'VocabularyService'
 ]

dashscope/audio/asr/recognition.py

@@ -1,7 +1,7 @@
 import json
 import os
-import time
 import threading
+import time
 from http import HTTPStatus
 from threading import Timer
 from typing import Any, Dict, List, Union
@@ -185,24 +185,29 @@ class Recognition(BaseApi):
                     self._callback.on_complete()
                 else:
                     usage: Dict[str, Any] = None
-                    useags: List[Any] = None
+                    usages: List[Any] = None
                     if 'sentence' in part.output:
                         if (self._first_package_timestamp < 0):
-                            self._first_package_timestamp = time.time()*1000
-                            logger.debug('first package delay {}'.format(self._first_package_timestamp - self._start_stream_timestamp))
+                            self._first_package_timestamp = time.time() * 1000
+                            logger.debug('first package delay {}'.format(
+                                self._first_package_timestamp -
+                                self._start_stream_timestamp))
                         sentence = part.output['sentence']
-                        logger.debug('Recv Result :{}, isEnd: {}'.format(sentence, RecognitionResult.is_sentence_end(sentence)))                        
-                        if part.usage is not None:                        
+                        logger.debug('Recv Result :{}, isEnd: {}'.format(
+                            sentence,
+                            RecognitionResult.is_sentence_end(sentence)))
+                        if part.usage is not None:
                             usage = {
-                                'end_time': part.output['sentence']['end_time'],
+                                'end_time':
+                                part.output['sentence']['end_time'],
                                 'usage': part.usage
                             }
-                            useags = [usage]
+                            usages = [usage]
 
                     self._callback.on_event(
                         RecognitionResult(
                             RecognitionResponse.from_api_response(part),
-                            usages=useags))
+                            usages=usages))
             else:
                 self._running = False
                 self._stream_data.clear()
@@ -318,7 +323,7 @@ class Recognition(BaseApi):
         Returns:
             RecognitionResult: The result of speech recognition.
         """
-        self._start_stream_timestamp = time.time()*1000
+        self._start_stream_timestamp = time.time() * 1000
         if self._running:
             raise InvalidParameter('Speech recognition has been called.')
 
@@ -363,10 +368,14 @@ class Recognition(BaseApi):
                 if part.status_code == HTTPStatus.OK:
                     if 'sentence' in part.output:
                         if (self._first_package_timestamp < 0):
-                            self._first_package_timestamp = time.time()*1000
-                            logger.debug('first package delay {}'.format(self._first_package_timestamp - self._start_stream_timestamp))                        
+                            self._first_package_timestamp = time.time() * 1000
+                            logger.debug('first package delay {}'.format(
+                                self._first_package_timestamp -
+                                self._start_stream_timestamp))
                         sentence = part.output['sentence']
-                        logger.debug('Recv Result :{}, isEnd: {}'.format(sentence, RecognitionResult.is_sentence_end(sentence)))
+                        logger.debug('Recv Result :{}, isEnd: {}'.format(
+                            sentence,
+                            RecognitionResult.is_sentence_end(sentence)))
                         if RecognitionResult.is_sentence_end(sentence):
                             sentences.append(sentence)
 

dashscope/audio/asr/transcribe.py

@@ -0,0 +1,270 @@
+import asyncio
+import os
+from http import HTTPStatus
+from typing import Any, Dict
+from urllib.parse import urlparse
+
+import aiohttp
+
+from dashscope.api_entities.dashscope_response import DashScopeAPIResponse
+from dashscope.client.base_api import BaseApi
+from dashscope.common.constants import ApiProtocol, HTTPMethod
+from dashscope.common.error import InputRequired
+from dashscope.common.utils import _get_task_group_and_task
+
+
+class Transcribe(BaseApi):
+    """API for File Transcriber models.
+
+    """
+
+    MAX_QUERY_TRY_COUNT = 3
+
+    @classmethod
+    def call(cls, model: str, file: str, **kwargs) -> DashScopeAPIResponse:
+        """Call file transcriber model service.
+
+        Args:
+            model (str): The requested model, such as paraformer-16k-1
+            file (str): The local path or URL of the file.
+            channel_id (List[int], optional): The selected channel_id of audio file. # noqa: E501
+
+        Returns:
+            DashScopeAPIResponse: The response body.
+
+        Raises:
+            InputRequired: The file cannot be empty.
+        """
+        loop = asyncio.new_event_loop()
+        asyncio.set_event_loop(loop)
+        return loop.run_until_complete(cls.async_call(model, file, **kwargs))
+
+    @classmethod
+    async def async_call(cls, model: str, file: str,
+                         **kwargs) -> DashScopeAPIResponse:
+        """Async call file transcriber model service.
+
+        Args:
+            model (str): The requested model, such as paraformer-16k-1
+            file (str): The local path or URL of the file.
+            channel_id (List[int], optional): The selected channel_id of audio file. # noqa: E501
+
+        Returns:
+            DashScopeAPIResponse: The response body.
+
+        Raises:
+            InputRequired: The file cannot be empty.
+        """
+        cls.is_url = cls._validate_file(file)
+        cls.file_name = file
+        cls.model_id = model
+
+        request = {'file': cls.file_name, 'is_url': cls.is_url}
+
+        # launch transcribe request, and get task info.
+        task = await cls._async_launch_requests(request, **kwargs)
+
+        response = await cls._async_get_result(task, **kwargs)
+
+        return response
+
+    @classmethod
+    async def _async_launch_requests(cls, request: Dict[str, Any], **kwargs):
+        """Async submit transcribe request.
+
+        Args:
+            inputs (Dict[str, Any]): The input parameters.
+
+        Returns:
+            task (Dict[str, Any]): The result of the task request.
+        """
+        inputs = {'file_link': request['file']}
+        task = {'file': request['file']}
+        local_file = None
+        try_count: int = 0
+        response = DashScopeAPIResponse(id='', code=HTTPStatus.OK, output=None)
+        if not request['is_url']:
+            try:
+                local_file = open(inputs['file_link'], 'rb')
+            except IOError as e:
+                raise InputRequired(f'File cannot be opened. {e}')
+
+            kwargs['form'] = {'av_file': local_file}
+
+        task_name, function = _get_task_group_and_task(__name__)
+        kwargs['async_request'] = True
+        kwargs['query'] = False
+
+        while True:
+            try:
+                response = await super().async_call(
+                    model=cls.model_id,
+                    task_group='audio',
+                    task=task_name,
+                    function=function,
+                    input=inputs,
+                    api_protocol=ApiProtocol.HTTP,
+                    http_method=HTTPMethod.POST,
+                    **kwargs)
+
+                task['request_id'] = response.id
+                task['code'] = response.code
+                task['status'] = response.status
+
+                if response.code == HTTPStatus.OK and response.output is not None:  # noqa: E501
+                    task.update(response.output)
+                else:
+                    task['message'] = response.message
+
+                break
+
+            except (asyncio.TimeoutError, aiohttp.ClientConnectorError) as e:
+                try_count += 1
+                if try_count > Transcribe.MAX_QUERY_TRY_COUNT:
+                    task['request_id'] = response.id
+                    task['code'] = HTTPStatus.REQUEST_TIMEOUT
+                    task['status'] = response.status
+                    task['message'] = str(e)
+                    break
+                else:
+                    await asyncio.sleep(2)
+                    continue
+            except Exception as e:
+                task['request_id'] = response.id
+                task['code'] = HTTPStatus.BAD_REQUEST
+                task['status'] = response.status
+                task['message'] = str(e)
+                break
+
+        if local_file is not None:
+            local_file.close()
+
+        return task
+
+    @classmethod
+    async def _async_get_result(cls, task, **kwargs):
+        """Async get transcribe result by polling.
+
+        Args:
+            task (Dict[str, Any]): The info of the task request.
+
+        Returns:
+            DashScopeAPIResponse: The response body.
+        """
+        request = task
+        responses = []
+        item = {}
+        response = DashScopeAPIResponse(id=request['request_id'],
+                                        code=request['code'],
+                                        output=None,
+                                        status=request['status'],
+                                        message=request['message'])
+
+        if request['code'] != HTTPStatus.OK:
+            item['file'] = request['file']
+            item['request_id'] = response.id
+            item['code'] = request['code']
+            item['status'] = request['status']
+            item['message'] = request['message']
+            responses.append(item)
+        else:
+            try_count: int = 0
+            while True:
+                item['file'] = request['file']
+                item['task_Id'] = request['task_id']
+
+                try:
+                    inputs = {}
+                    inputs['task_Id'] = request['task_id']
+                    kwargs['async_request'] = True
+                    kwargs['query'] = True
+
+                    response = await super().async_call(
+                        model=cls.model_id,
+                        task_group=None,
+                        task='tasks',
+                        input=inputs,
+                        task_id=inputs['task_Id'],
+                        api_protocol=ApiProtocol.HTTP,
+                        http_method=HTTPMethod.GET,
+                        **kwargs)
+                except (asyncio.TimeoutError,
+                        aiohttp.ClientConnectorError) as e:
+                    try_count += 1
+                    if try_count > Transcribe.MAX_QUERY_TRY_COUNT:
+                        item['request_id'] = response.id
+                        item['code'] = HTTPStatus.REQUEST_TIMEOUT
+                        item['status'] = response.status
+                        item['message'] = str(e)
+                        responses.append(item)
+                        break
+                    else:
+                        await asyncio.sleep(2)
+                        continue
+                except Exception as e:
+                    item['request_id'] = response.id
+                    item['code'] = HTTPStatus.BAD_REQUEST
+                    item['status'] = response.status
+                    item['message'] = str(e)
+                    responses.append(item)
+                    break
+
+                try_count = 0
+                item['request_id'] = response.id
+                item['code'] = response.code
+                item['status'] = response.status
+
+                if response.code == HTTPStatus.OK:
+                    if 'task_status' in response.output:
+                        task_status = response.output['task_status']
+                        if task_status == 'QUEUING' or task_status == 'PROCESSING':  # noqa: E501
+                            await asyncio.sleep(2)
+                            continue
+
+                    item.update(response.output)
+                else:
+                    item['message'] = response.message
+
+                responses.append(item)
+                break
+
+        output = {}
+        output['results'] = responses
+
+        return DashScopeAPIResponse(id=response.id,
+                                    code=response.code,
+                                    status=response.status,
+                                    message=response.message,
+                                    output=output)
+
+    @classmethod
+    def _validate_file(cls, file: str):
+        """Check the validity of the file
+        and whether the file is a URL or a local path.
+
+        Args:
+            file (str): The local path or URL of the file.
+
+        Returns:
+            bool: Whether the file is a URL.
+        """
+        if file is None or len(file) == 0:
+            raise InputRequired(
+                'Input an illegal file, please ensure that the file type is a local path or URL!'  # noqa: *
+            )
+
+        if os.path.isfile(file):
+            return False
+        else:
+            result = urlparse(file)
+            if result.scheme is not None and len(result.scheme) > 0:
+                if result.scheme != 'http' and result.scheme != 'https':
+                    raise InputRequired(
+                        f'The URL protocol({result.scheme}) of file({file}) is not http or https.'  # noqa: *
+                    )
+            else:
+                raise InputRequired(
+                    f'Input an illegal file({file}), maybe the file is inexistent.'  # noqa: *
+                )
+
+        return True

dashscope/audio/asr/transcription.py

@@ -3,7 +3,6 @@ import time
 from typing import List, Union
 
 import aiohttp
-
 from dashscope.api_entities.dashscope_response import (DashScopeAPIResponse,
                                                        TranscriptionResponse)
 from dashscope.client.base_api import BaseAsyncApi
@@ -109,13 +108,11 @@ class Transcription(BaseAsyncApi):
         return TranscriptionResponse.from_api_response(response)
 
     @classmethod
-    def fetch(
-        cls,
-        task: Union[str, TranscriptionResponse],
-        api_key: str = None,
-        workspace: str = None,
-        **kwargs
-    ) -> TranscriptionResponse:
+    def fetch(cls,
+              task: Union[str, TranscriptionResponse],
+              api_key: str = None,
+              workspace: str = None,
+              **kwargs) -> TranscriptionResponse:
         """Fetch the status of task, including results of batch transcription when task_status is SUCCEEDED.  # noqa: E501
 
         Args:
@@ -147,13 +144,11 @@ class Transcription(BaseAsyncApi):
         return TranscriptionResponse.from_api_response(response)
 
     @classmethod
-    def wait(
-        cls,
-        task: Union[str, TranscriptionResponse],
-        api_key: str = None,
-        workspace: str = None,
-        **kwargs
-    ) -> TranscriptionResponse:
+    def wait(cls,
+             task: Union[str, TranscriptionResponse],
+             api_key: str = None,
+             workspace: str = None,
+             **kwargs) -> TranscriptionResponse:
         """Poll task until the final results of transcription is obtained.
 
         Args:
@@ -164,7 +159,10 @@ class Transcription(BaseAsyncApi):
         Returns:
             TranscriptionResponse: The result of batch transcription.
         """
-        response = super().wait(task, api_key=api_key, workspace=workspace, **kwargs)
+        response = super().wait(task,
+                                api_key=api_key,
+                                workspace=workspace,
+                                **kwargs)
         return TranscriptionResponse.from_api_response(response)
 
     @classmethod

dashscope/audio/asr/vocabulary.py

@@ -0,0 +1,173 @@
+import asyncio
+import time
+from typing import List
+
+import aiohttp
+from dashscope.client.base_api import BaseApi
+from dashscope.common.constants import ApiProtocol, HTTPMethod
+from dashscope.common.logging import logger
+
+
+class VocabularyServiceException(Exception):
+    def __init__(self, status_code: int, code: str,
+                 error_message: str) -> None:
+        self._status_code = status_code
+        self._code = code
+        self._error_message = error_message
+
+    def __str__(self):
+        return f'Status Code: {self._status_code}, Code: {self._code}, Error Message: {self._error_message}'
+
+
+class VocabularyService(BaseApi):
+    '''
+    API for asr vocabulary service
+    '''
+    MAX_QUERY_TRY_COUNT = 3
+
+    def __init__(self,
+                 api_key=None,
+                 workspace=None,
+                 model=None,
+                 **kwargs) -> None:
+        super().__init__()
+        self._api_key = api_key
+        self._workspace = workspace
+        self._kwargs = kwargs
+        self._last_request_id = None
+        self.model = model
+        if self.model == None:
+            self.model = 'speech-biasing'
+
+    def __call_with_input(self, input):
+        try_count = 0        
+        while True:
+            try:
+                response = super().call(model=self.model,
+                                        task_group='audio',
+                                        task='asr',
+                                        function='customization',
+                                        input=input,
+                                        api_protocol=ApiProtocol.HTTP,
+                                        http_method=HTTPMethod.POST,
+                                        api_key=self._api_key,
+                                        workspace=self._workspace,
+                                        **self._kwargs)
+            except (asyncio.TimeoutError, aiohttp.ClientConnectorError) as e:
+                logger.error(e)
+                try_count += 1
+                if try_count <= VocabularyService.MAX_QUERY_TRY_COUNT:
+                    time.sleep(2)
+                    continue
+
+            break
+        logger.debug('>>>>recv', response)
+        return response
+
+    def create_vocabulary(self, target_model: str, prefix: str,
+                          vocabulary: List[dict]) -> str:
+        '''
+        创建热词表
+        param: target_model 热词表对应的语音识别模型版本
+        param: prefix 热词表自定义前缀，仅允许数字和小写字母，小于十个字符。
+        param: vocabulary 热词表字典
+        return: 热词表标识符 vocabulary_id
+        '''
+        response = self.__call_with_input(input={
+            'action': 'create_vocabulary',
+            'target_model': target_model,
+            'prefix': prefix,
+            'vocabulary': vocabulary,
+        }, )
+        if response.status_code == 200:
+            self._last_request_id = response.request_id
+            return response.output['vocabulary_id']
+        else:
+            raise VocabularyServiceException(response.status_code,
+                                             response.code, response.message)
+
+    def list_vocabularies(self,
+                          prefix=None,
+                          page_index: int = 0,
+                          page_size: int = 10) -> List[dict]:
+        '''
+        查询已创建的所有热词表
+        param: prefix 自定义前缀，如果设定则只返回指定前缀的热词表标识符列表。
+        param: page_index 查询的页索引
+        param: page_size 查询页大小
+        return: 热词表标识符列表
+        '''
+        if prefix:
+            response = self.__call_with_input(input={
+                'action': 'list_vocabulary',
+                'prefix': prefix,
+                'page_index': page_index,
+                'page_size': page_size,
+            }, )
+        else:
+            response = self.__call_with_input(input={
+                'action': 'list_vocabulary',
+                'page_index': page_index,
+                'page_size': page_size,
+            }, )
+        if response.status_code == 200:
+            self._last_request_id = response.request_id
+            return response.output['vocabulary_list']
+        else:
+            raise VocabularyServiceException(response.status_code,
+                                             response.code, response.message)
+
+    def query_vocabulary(self, vocabulary_id: str) -> List[dict]:
+        '''
+        获取热词表内容
+        param: vocabulary_id 热词表标识符
+        return: 热词表
+        '''
+        response = self.__call_with_input(input={
+            'action': 'query_vocabulary',
+            'vocabulary_id': vocabulary_id,
+        }, )
+        if response.status_code == 200:
+            self._last_request_id = response.request_id
+            return response.output
+        else:
+            raise VocabularyServiceException(response.status_code,
+                                             response.code, response.message)
+
+    def update_vocabulary(self, vocabulary_id: str,
+                          vocabulary: List[dict]) -> None:
+        '''
+        用新的热词表替换已有热词表
+        param: vocabulary_id 需要替换的热词表标识符
+        param: vocabulary 热词表
+        '''
+        response = self.__call_with_input(input={
+            'action': 'update_vocabulary',
+            'vocabulary_id': vocabulary_id,
+            'vocabulary': vocabulary,
+        }, )
+        if response.status_code == 200:
+            self._last_request_id = response.request_id
+            return
+        else:
+            raise VocabularyServiceException(response.status_code,
+                                             response.code, response.message)
+
+    def delete_vocabulary(self, vocabulary_id: str) -> None:
+        '''
+        删除热词表
+        param: vocabulary_id 需要删除的热词表标识符
+        '''
+        response = self.__call_with_input(input={
+            'action': 'delete_vocabulary',
+            'vocabulary_id': vocabulary_id,
+        }, )
+        if response.status_code == 200:
+            self._last_request_id = response.request_id
+            return
+        else:
+            raise VocabularyServiceException(response.status_code,
+                                             response.code, response.message)
+
+    def get_last_request_id(self):
+        return self._last_request_id

dashscope/audio/tts_v2/__init__.py

@@ -1,7 +1,7 @@
+from .enrollment import VoiceEnrollmentException, VoiceEnrollmentService
 from .speech_synthesizer import AudioFormat, ResultCallback, SpeechSynthesizer
 
-__all__ = ['SpeechSynthesizer', 'ResultCallback', 'AudioFormat']
-
-# from .speech_synthesizer import (SpeechSynthesizer, ResultCallback, SpeechSynthesisResult, AudioFormat)
-
-# __all__ = ['SpeechSynthesizer', 'ResultCallback', 'SpeechSynthesisResult', 'AudioFormat']
+__all__ = [
+    'SpeechSynthesizer', 'ResultCallback', 'AudioFormat',
+    'VoiceEnrollmentException', 'VoiceEnrollmentService'
+]

dashscope/audio/tts_v2/enrollment.py

@@ -0,0 +1,170 @@
+import asyncio
+import time
+from typing import List
+
+import aiohttp
+from dashscope.client.base_api import BaseApi
+from dashscope.common.constants import ApiProtocol, HTTPMethod
+from dashscope.common.logging import logger
+
+
+class VoiceEnrollmentException(Exception):
+    def __init__(self, status_code: int, code: str,
+                 error_message: str) -> None:
+        self._status_code = status_code
+        self._code = code
+        self._error_message = error_message
+
+    def __str__(self):
+        return f'Status Code: {self._status_code}, Code: {self._code}, Error Message: {self._error_message}'
+
+
+class VoiceEnrollmentService(BaseApi):
+    '''
+    API for voice clone service
+    '''
+    MAX_QUERY_TRY_COUNT = 3
+
+    def __init__(self,
+                 api_key=None,
+                 workspace=None,
+                 model=None,
+                 **kwargs) -> None:
+        super().__init__()
+        self._api_key = api_key
+        self._workspace = workspace
+        self._kwargs = kwargs
+        self._last_request_id = None
+        self.model = model
+        if self.model == None:
+            self.model = 'voice-enrollment'
+
+    def __call_with_input(self, input):
+        try_count = 0
+        while True:
+            try:
+                response = super().call(model=self.model,
+                                        task_group='audio',
+                                        task='tts',
+                                        function='customization',
+                                        input=input,
+                                        api_protocol=ApiProtocol.HTTP,
+                                        http_method=HTTPMethod.POST,
+                                        api_key=self._api_key,
+                                        workspace=self._workspace,
+                                        **self._kwargs)
+            except (asyncio.TimeoutError, aiohttp.ClientConnectorError) as e:
+                logger.error(e)
+                try_count += 1
+                if try_count <= VoiceEnrollmentService.MAX_QUERY_TRY_COUNT:
+                    time.sleep(2)
+                    continue
+
+            break
+        logger.debug('>>>>recv', response)
+        return response
+
+    def create_voice(self, target_model: str, prefix: str, url: str) -> str:
+        '''
+        创建新克隆音色
+        param: target_model 克隆音色对应的语音识别模型版本
+        param: prefix 音色自定义前缀，仅允许数字和小写字母，小于十个字符。
+        param: url 用于克隆的音频文件url
+        return: voice_id
+        '''
+        response = self.__call_with_input(input={
+            'action': 'create_voice',
+            'target_model': target_model,
+            'prefix': prefix,
+            'url': url,
+        }, )
+        if response.status_code == 200:
+            self._last_request_id = response.request_id
+            return response.output['voice_id']
+        else:
+            raise VoiceEnrollmentException(response.status_code, response.code,
+                                           response.message)
+
+    def list_voices(self,
+                    prefix=None,
+                    page_index: int = 0,
+                    page_size: int = 10) -> List[dict]:
+        '''
+        查询已创建的所有音色
+        param: page_index 查询的页索引
+        param: page_size 查询页大小
+        return: List[dict] 音色列表，包含每个音色的id，创建时间，修改时间，状态。
+        '''
+        if prefix:
+            response = self.__call_with_input(input={
+                'action': 'list_voice',
+                'prefix': prefix,
+                'page_index': page_index,
+                'page_size': page_size,
+            }, )
+        else:
+            response = self.__call_with_input(input={
+                'action': 'list_voice',
+                'page_index': page_index,
+                'page_size': page_size,
+            }, )
+        if response.status_code == 200:
+            self._last_request_id = response.request_id
+            return response.output['voice_list']
+        else:
+            raise VoiceEnrollmentException(response.status_code, response.code,
+                                           response.message)
+
+    def query_voice(self, voice_id: str) -> List[str]:
+        '''
+        查询已创建的所有音色
+        param: voice_id 需要查询的音色
+        return: bytes 注册音色使用的音频
+        '''
+        response = self.__call_with_input(input={
+            'action': 'query_voice',
+            'voice_id': voice_id,
+        }, )
+        if response.status_code == 200:
+            self._last_request_id = response.request_id
+            return response.output
+        else:
+            raise VoiceEnrollmentException(response.status_code, response.code,
+                                           response.message)
+
+    def update_voice(self, voice_id: str, url: str) -> None:
+        '''
+        更新音色
+        param: voice_id 音色id
+        param: url 用于克隆的音频文件url
+        '''
+        response = self.__call_with_input(input={
+            'action': 'update_voice',
+            'voice_id': voice_id,
+            'url': url,
+        }, )
+        if response.status_code == 200:
+            self._last_request_id = response.request_id
+            return
+        else:
+            raise VoiceEnrollmentException(response.status_code, response.code,
+                                           response.message)
+
+    def delete_voice(self, voice_id: str) -> None:
+        '''
+        删除音色
+        param: voice_id 需要删除的音色
+        '''
+        response = self.__call_with_input(input={
+            'action': 'delete_voice',
+            'voice_id': voice_id,
+        }, )
+        if response.status_code == 200:
+            self._last_request_id = response.request_id
+            return
+        else:
+            raise VoiceEnrollmentException(response.status_code, response.code,
+                                           response.message)
+
+    def get_last_request_id(self):
+        return self._last_request_id

dashscope/audio/tts_v2/speech_synthesizer.py

@@ -80,7 +80,7 @@ class Request:
         voice,
         format='wav',
         sample_rate=16000,
-        volumn=50,
+        volume=50,
         speech_rate=1.0,
         pitch_rate=1.0,
     ):
@@ -90,7 +90,7 @@ class Request:
         self.model = model
         self.format = format
         self.sample_rate = sample_rate
-        self.volumn = volumn
+        self.volume = volume
         self.speech_rate = speech_rate
         self.pitch_rate = pitch_rate
 
@@ -136,7 +136,7 @@ class Request:
                 },
                 'parameters': {
                     'voice': self.voice,
-                    'volume': self.volumn,
+                    'volume': self.volume,
                     'text_type': 'PlainText',
                     'sample_rate': self.sample_rate,
                     'rate': self.speech_rate,
@@ -190,7 +190,7 @@ class SpeechSynthesizer:
         model,
         voice,
         format: AudioFormat = AudioFormat.DEFAULT,
-        volumn=50,
+        volume=50,
         speech_rate=1.0,
         pitch_rate=1.0,
         headers=None,
@@ -253,7 +253,7 @@ class SpeechSynthesizer:
             voice=voice,
             format=format.format,
             sample_rate=format.sample_rate,
-            volumn=volumn,
+            volume=volume,
             speech_rate=speech_rate,
             pitch_rate=pitch_rate,
         )
@@ -424,6 +424,8 @@ class SpeechSynthesizer:
         request = self.request.getFinishRequest()
         self.__send_str(request)
         self.close()
+        self.start_event.set()
+        self.complete_event.set()
 
     # 监听消息的回调函数
     def on_message(self, ws, message):

dashscope/threads/runs/runs.py

@@ -386,12 +386,6 @@ class Runs(CreateMixin, CancelMixin, ListObjectMixin, GetStatusMixin,
                           thread_id=thread_id,
                           workspace=workspace,
                           api_key=api_key)
-            import json
-            print(
-                json.dumps(run,
-                           default=lambda o: o.__dict__,
-                           sort_keys=True,
-                           indent=4))
             if run.status_code == HTTPStatus.OK:
                 if hasattr(run, 'status'):
                     if run.status in [

dashscope/utils/oss_utils.py

@@ -121,8 +121,51 @@ def upload_file(model: str, upload_path: str, api_key: str):
     return None
 
 
+def check_and_upload_local(model: str, content: str, api_key: str):
+    """Check the content is local file path, upload and return the url
+
+    Args:
+        model (str): Which model to upload.
+        content (str): The content.
+        api_key (_type_): The api key.
+
+    Raises:
+        UploadFileException: Upload failed.
+        InvalidInput: The input is invalid
+
+    Returns:
+        _type_: if upload return True and file_url otherwise False, origin content.
+    """
+    if content.startswith(FILE_PATH_SCHEMA):
+        parse_result = urlparse(content)
+        if parse_result.netloc:
+            file_path = parse_result.netloc + unquote_plus(parse_result.path)
+        else:
+            file_path = unquote_plus(parse_result.path)
+        if os.path.exists(file_path):
+            file_url = OssUtils.upload(model=model,
+                                       file_path=file_path,
+                                       api_key=api_key)
+            if file_url is None:
+                raise UploadFileException('Uploading file: %s failed' %
+                                          content)
+            return True, file_url
+        else:
+            raise InvalidInput('The file: %s is not exists!' % file_path)
+    elif not content.startswith('http'):
+        if os.path.exists(content):
+            file_url = OssUtils.upload(model=model,
+                                       file_path=content,
+                                       api_key=api_key)
+            if file_url is None:
+                raise UploadFileException('Uploading file: %s failed' %
+                                          content)
+            return True, file_url
+    return False, content
+
+
 def check_and_upload(model, elem: dict, api_key):
-    is_upload = False
+    has_upload = False
     for key, content in elem.items():
         # support video:[images] for qwen2-vl
         is_list = isinstance(content, list)
@@ -130,38 +173,14 @@ def check_and_upload(model, elem: dict, api_key):
 
         if key in ['image', 'video', 'audio', 'text']:
             for i, content in enumerate(contents):
-                if content.startswith(FILE_PATH_SCHEMA):
-                    parse_result = urlparse(content)
-                    if parse_result.netloc:
-                        file_path = parse_result.netloc + unquote_plus(
-                            parse_result.path)
-                    else:
-                        file_path = unquote_plus(parse_result.path)
-                    if os.path.exists(file_path):
-                        file_url = OssUtils.upload(model=model,
-                                                   file_path=file_path,
-                                                   api_key=api_key)
-                        if file_url is None:
-                            raise UploadFileException(
-                                'Uploading file: %s failed' % content)
-                        contents[i] = file_url
-                        is_upload = True
-                    else:
-                        raise InvalidInput('The file: %s is not exists!' %
-                                           file_path)
-                elif not content.startswith('http'):
-                    if os.path.exists(content):
-                        file_url = OssUtils.upload(model=model,
-                                                   file_path=content,
-                                                   api_key=api_key)
-                        if file_url is None:
-                            raise UploadFileException(
-                                'Uploading file: %s failed' % content)
-                        contents[i] = file_url
-                        is_upload = True
+                is_upload, file_url = check_and_upload_local(
+                    model, content, api_key)
+                if is_upload:
+                    contents[i] = file_url
+                    has_upload = True
         elem[key] = contents if is_list else contents[0]
 
-    return is_upload
+    return has_upload
 
 
 def preprocess_message_element(model: str, elem: List[dict], api_key: str):

dashscope/version.py

@@ -1 +1 @@
-__version__ = '1.20.8'
+__version__ = '1.20.10'

{dashscope-1.20.8.dist-info → dashscope-1.20.10.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: dashscope
-Version: 1.20.8
+Version: 1.20.10
 Summary: dashscope client sdk library
 Home-page: https://dashscope.aliyun.com/
 Author: Alibaba Cloud

{dashscope-1.20.8.dist-info → dashscope-1.20.10.dist-info}/RECORD

@@ -6,12 +6,12 @@ dashscope/files.py,sha256=QgJjwhtn9F548nCA8jD8OvE6aQEj-20hZqJgYXsUdQU,3930
 dashscope/finetune.py,sha256=_tflDUvu0KagSoCzLaf0hofpG_P8NU6PylL8CPjVhrA,6243
 dashscope/model.py,sha256=UPOn1qMYFhX-ovXi3BMxZEBk8qOK7WLJOYHMbPZwYBo,1440
 dashscope/models.py,sha256=1-bc-Ue68zurgu_y6RhfFr9uzeQMF5AZq-C32lJGMGU,1224
-dashscope/version.py,sha256=528qtkrW31Rybnpunug78DmLVf7Ys8Mn-aRU-fFyAdA,23
+dashscope/version.py,sha256=9ivfT91xguUu9PULHzh1-7jFzb5sdLCuzbdRnbN8EGo,24
 dashscope/aigc/__init__.py,sha256=s-MCA87KYiVumYtKtJi5IMN7xelSF6TqEU3s3_7RF-Y,327
 dashscope/aigc/code_generation.py,sha256=KAJVrGp6tiNFBBg64Ovs9RfcP5SrIhrbW3wdA89NKso,10885
 dashscope/aigc/conversation.py,sha256=xRoJlCR-IXHjSdkDrK74a9ut1FJg0FZhTNXZAJC18MA,14231
 dashscope/aigc/generation.py,sha256=53oMCmN5ZbqeqAsKxmdunXlRh-XP8ZtnA7hB2id4Koo,17897
-dashscope/aigc/image_synthesis.py,sha256=_TGkh4L_yBNQNmoJUKlfXfljKfxX-SSyxHHQBDm1AC8,10418
+dashscope/aigc/image_synthesis.py,sha256=UWHW-nvf7_aDZKr4uZDusVHjqWr9TSZjCsZI8YSWaek,11052
 dashscope/aigc/multimodal_conversation.py,sha256=SlNnnsUPV19gdx8fYJAtsMFWPNGY6vhk5IGHZ5ZczpI,5369
 dashscope/api_entities/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 dashscope/api_entities/aiohttp_request.py,sha256=aE3AeWba8Ig_xHMYjrAdkq0N61l_L2VFTG6HYh912X0,10229
@@ -29,14 +29,17 @@ dashscope/assistants/assistant_types.py,sha256=1jNL30TOlrkiYhvCaB3E8jkPLG8CnQ6I3
 dashscope/assistants/assistants.py,sha256=NYahIDqhtnOcQOmnhZsjc5F5jvBUQcce8-fbrJXHVnQ,10833
 dashscope/assistants/files.py,sha256=pwLVJ_pjpRFWyfI_MRvhH7Si7FzGDj4ChzZgWTpLOhg,6699
 dashscope/audio/__init__.py,sha256=-ZRxrK-gV4QsUtlThIT-XwqB6vmyEsnhxIxdLmhCUuc,61
-dashscope/audio/asr/__init__.py,sha256=-s180qWn_JPSpCo1q0aDJJ5HQ3zTzD4z5yUwsRqH4aU,275
+dashscope/audio/asr/__init__.py,sha256=kFdx3IYsdfGGDDlQmUjvtd2kqifuEekwlPBEOUvXvEY,406
 dashscope/audio/asr/asr_phrase_manager.py,sha256=EjtbI3zz9UQGS1qv6Yb4zzEMj4OJJVXmwkqZyIrzvEA,7642
-dashscope/audio/asr/recognition.py,sha256=cEooE3wGf8kKfJIVbaXEytl5X6F0hMsLe8g4Bj9Fn4w,18768
-dashscope/audio/asr/transcription.py,sha256=1WAg9WH89antVzRYEKXb5LQP9xylZmX4YKp7v5oMYjY,8931
+dashscope/audio/asr/recognition.py,sha256=a4zIkIMiWwOEApP9k9ZC9jGDr7CP7BqB6Cy1dBVTN4g,18978
+dashscope/audio/asr/transcribe.py,sha256=HfZYpvpVfvGRAIIIzX65Af33E6vsIFGd_qqhQ8LaNcM,9651
+dashscope/audio/asr/transcription.py,sha256=D8CW0XDqJuEJVmNFJ6qczTysSV3Sz_rzk2C6NIKTtVc,9042
+dashscope/audio/asr/vocabulary.py,sha256=880u5CGh8Ky9iWXDf_7cUuHfL5AGmw8JJRCbRThVCMI,6484
 dashscope/audio/tts/__init__.py,sha256=fbnieZX9yNFNh5BsxLpLXb63jlxzxrdCJakV3ignjlQ,194
 dashscope/audio/tts/speech_synthesizer.py,sha256=dnKx9FDDdO_ETHAjhK8zaMVaH6SfoTtN5YxXXqgY1JA,7571
-dashscope/audio/tts_v2/__init__.py,sha256=ve5a81qTbWDcRaSuritZtJBzryOIol2_dxzfqqdCw-k,345
-dashscope/audio/tts_v2/speech_synthesizer.py,sha256=bpzj9gx2D_FfOzgsjU-GBGmeWvEdewNPFd447mOgM-o,19220
+dashscope/audio/tts_v2/__init__.py,sha256=5UfyDBYnuGgOy9KMxEIXA2U2ihcXutdZc1cqJudy-8M,282
+dashscope/audio/tts_v2/enrollment.py,sha256=sUkOEUsP8RXREMtTkAeDTYfrQJ6lPnM_Y-DeefXB_Q4,6140
+dashscope/audio/tts_v2/speech_synthesizer.py,sha256=lATasQJB8HlB_yYm90qqW6zIAE1CQFxBxhnch6xdg9s,19285
 dashscope/client/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 dashscope/client/base_api.py,sha256=rXN97XGyDhCCaD_dz_clpFDjOJfpGjqiH7yX3LaD-GE,41233
 dashscope/common/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -73,7 +76,7 @@ dashscope/threads/messages/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJ
 dashscope/threads/messages/files.py,sha256=wi0nJ2zsPWOw2Jn-ZkxA3URZBIrkGxqM_uAPfXY1xv0,3820
 dashscope/threads/messages/messages.py,sha256=Zjmyf3rT1XSdn33hPrqOY6DSWUVL7pDEapG03FREPV8,8419
 dashscope/threads/runs/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-dashscope/threads/runs/runs.py,sha256=Cvy5FD0x1Z9c5qayYeNpoL_QIqH4yxgqdGplCk3soRw,18762
+dashscope/threads/runs/runs.py,sha256=ux4VH_lxxHCw1XOqngzmsm9kwTR3jS0wX27xoAswHlY,18549
 dashscope/threads/runs/steps.py,sha256=pLNR-5g7zvYkvC-p4sZGVgYHd1jqxBerM2WFyB358H8,3638
 dashscope/tokenizers/__init__.py,sha256=Oy5FMT37Non6e1YxdHQ89U93Dy3CG1Ez0gBa771KZo0,200
 dashscope/tokenizers/qwen_tokenizer.py,sha256=dCnT9-9NrqPS85bEhjlPULUfDADVRhlleYwM_ILgCeI,4111
@@ -81,10 +84,10 @@ dashscope/tokenizers/tokenization.py,sha256=G6cSEmVLr3pjXUC3EOU9ot8MYxNnOQ4wOB2m
 dashscope/tokenizers/tokenizer.py,sha256=y6P91qTCYo__pEx_0VHAcj9YECfbUdRqZU1fdGTjF4o,1154
 dashscope/tokenizers/tokenizer_base.py,sha256=REDhzRyDT13iequ61-a6_KcTy0GFKlihQve5HkyoyRs,656
 dashscope/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-dashscope/utils/oss_utils.py,sha256=TSBh7MJK4ZH40Mxd8wNEsG8nQLNhQjWRjW3itHsvoZ0,7023
-dashscope-1.20.8.dist-info/LICENSE,sha256=Izp5L1DF1Mbza6qojkqNNWlE_mYLnr4rmzx2EBF8YFw,11413
-dashscope-1.20.8.dist-info/METADATA,sha256=M4hx1_1AZmfHDck7Aw4tYEKT4Fe3Ogf_iXFlIGU_cBc,6641
-dashscope-1.20.8.dist-info/WHEEL,sha256=G16H4A3IeoQmnOrYV4ueZGKSjhipXx8zc8nu9FGlvMA,92
-dashscope-1.20.8.dist-info/entry_points.txt,sha256=raEp5dOuj8whJ7yqZlDM8WQ5p2RfnGrGNo0QLQEnatY,50
-dashscope-1.20.8.dist-info/top_level.txt,sha256=woqavFJK9zas5xTqynmALqOtlafghjsk63Xk86powTU,10
-dashscope-1.20.8.dist-info/RECORD,,
+dashscope/utils/oss_utils.py,sha256=7vZ2Lypxwiit8VcAqAvr3cCyhVfaLapDiNuF-H3ZCD4,7332
+dashscope-1.20.10.dist-info/LICENSE,sha256=Izp5L1DF1Mbza6qojkqNNWlE_mYLnr4rmzx2EBF8YFw,11413
+dashscope-1.20.10.dist-info/METADATA,sha256=tjvuGjHvDHLCC1spvuhLM9GF_ImmIffIGifJAFdya3s,6642
+dashscope-1.20.10.dist-info/WHEEL,sha256=G16H4A3IeoQmnOrYV4ueZGKSjhipXx8zc8nu9FGlvMA,92
+dashscope-1.20.10.dist-info/entry_points.txt,sha256=raEp5dOuj8whJ7yqZlDM8WQ5p2RfnGrGNo0QLQEnatY,50
+dashscope-1.20.10.dist-info/top_level.txt,sha256=woqavFJK9zas5xTqynmALqOtlafghjsk63Xk86powTU,10
+dashscope-1.20.10.dist-info/RECORD,,