dashscope 1.24.5__py3-none-any.whl → 1.24.7__py3-none-any.whl

dashscope/aigc/image_synthesis.py

@@ -138,8 +138,15 @@ class ImageSynthesis(BaseAsyncApi):
         has_upload = False
         if negative_prompt is not None:
             inputs[NEGATIVE_PROMPT] = negative_prompt
-        if images is not None:
-            inputs[IMAGES] = images
+        if images is not None and images and len(images) > 0:
+            new_images = []
+            for image in images:
+                is_upload, new_image = check_and_upload_local(
+                    model, image, api_key)
+                if is_upload:
+                    has_upload = True
+                new_images.append(new_image)
+            inputs[IMAGES] = new_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)
@@ -178,10 +185,20 @@ class ImageSynthesis(BaseAsyncApi):
             headers['X-DashScope-OssResourceResolve'] = 'enable'
             kwargs['headers'] = headers
 
-        if task is None:
-            task = ImageSynthesis.task
-        if model is not None and model and 'imageedit' in model:
-            task = 'image2image'
+        def __get_i2i_task(task, model) -> str:
+            # 处理task参数：优先使用有效的task值
+            if task is not None and task != "":
+                return task
+
+            # 根据model确定任务类型
+            if model is not None and model != "":
+                if 'imageedit' in model or "wan2.5-i2i" in model:
+                    return 'image2image'
+
+            # 默认返回文本到图像任务
+            return ImageSynthesis.task
+
+        task = __get_i2i_task(task, model)
 
         return inputs, kwargs, task
 

dashscope/aigc/multimodal_conversation.py

@@ -28,6 +28,8 @@ class MultiModalConversation(BaseApi):
         api_key: str = None,
         workspace: str = None,
         text: str = None,
+        voice: str = None,
+        language_type: str = None,
         **kwargs
     ) -> Union[MultiModalConversationResponse, Generator[
             MultiModalConversationResponse, None, None]]:
@@ -57,6 +59,9 @@ class MultiModalConversation(BaseApi):
                 [1]: https://help.aliyun.com/zh/dashscope/developer-reference/api-key-settings. # noqa E501
             workspace (str): The dashscope workspace id.
             text (str): The text to generate.
+            voice (str): The voice name of qwen tts, include 'Cherry'/'Ethan'/'Sunny'/'Dylan' and so on,
+                    you can get the total voice list : https://help.aliyun.com/zh/model-studio/qwen-tts.
+            language_type (str): The synthesized language type, default is 'auto', useful for [qwen3-tts].
             **kwargs:
                 stream(bool, `optional`): Enable server-sent events
                     (ref: https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events)  # noqa E501
@@ -70,8 +75,6 @@ class MultiModalConversation(BaseApi):
                     tokens with top_p probability mass. So 0.1 means only
                     the tokens comprising the top 10% probability mass are
                     considered[qwen-turbo,bailian-v1].
-                voice(string, `optional`): The voice name of qwen tts, include 'Cherry'/'Ethan'/'Sunny'/'Dylan' and so on,
-                    you can get the total voice list : https://help.aliyun.com/zh/model-studio/qwen-tts.
                 top_k(float, `optional`):
 
 
@@ -99,6 +102,10 @@ class MultiModalConversation(BaseApi):
 
         if text is not None and text:
             input.update({'text': text})
+        if voice is not None and voice:
+            input.update({'voice': voice})
+        if language_type is not None and language_type:
+            input.update({'language_type': language_type})
         if msg_copy is not None:
             input.update({'messages': msg_copy})
         response = super().call(model=model,
@@ -160,6 +167,8 @@ class AioMultiModalConversation(BaseAioApi):
         api_key: str = None,
         workspace: str = None,
         text: str = None,
+        voice: str = None,
+        language_type: str = None,
         **kwargs
     ) -> Union[MultiModalConversationResponse, Generator[
             MultiModalConversationResponse, None, None]]:
@@ -189,6 +198,9 @@ class AioMultiModalConversation(BaseAioApi):
                 [1]: https://help.aliyun.com/zh/dashscope/developer-reference/api-key-settings. # noqa E501
             workspace (str): The dashscope workspace id.
             text (str): The text to generate.
+            voice (str): The voice name of qwen tts, include 'Cherry'/'Ethan'/'Sunny'/'Dylan' and so on,
+                    you can get the total voice list : https://help.aliyun.com/zh/model-studio/qwen-tts.
+            language_type (str): The synthesized language type, default is 'auto', useful for [qwen3-tts].
             **kwargs:
                 stream(bool, `optional`): Enable server-sent events
                     (ref: https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events)  # noqa E501
@@ -202,8 +214,6 @@ class AioMultiModalConversation(BaseAioApi):
                     tokens with top_p probability mass. So 0.1 means only
                     the tokens comprising the top 10% probability mass are
                     considered[qwen-turbo,bailian-v1].
-                voice(string, `optional`): The voice name of qwen tts, include 'Cherry'/'Ethan'/'Sunny'/'Dylan' and so on,
-                    you can get the total voice list : https://help.aliyun.com/zh/model-studio/qwen-tts.
                 top_k(float, `optional`):
 
         Raises:
@@ -230,6 +240,10 @@ class AioMultiModalConversation(BaseAioApi):
 
         if text is not None and text:
             input.update({'text': text})
+        if voice is not None and voice:
+            input.update({'voice': voice})
+        if language_type is not None and language_type:
+            input.update({'language_type': language_type})
         if msg_copy is not None:
             input.update({'messages': msg_copy})
         response = await super().call(model=model,

dashscope/aigc/video_synthesis.py

@@ -38,6 +38,7 @@ class VideoSynthesis(BaseAsyncApi):
              negative_prompt: str = None,
              template: str = None,
              img_url: str = None,
+             audio_url: str = None,
              api_key: str = None,
              extra_input: Dict = None,
              workspace: str = None,
@@ -56,6 +57,7 @@ class VideoSynthesis(BaseAsyncApi):
             negative_prompt (str): The negative prompt is the opposite of the prompt meaning.
             template (str): LoRa input, such as gufeng, katong, etc.
             img_url (str): The input image url, Generate the URL of the image referenced by the video.
+            audio_url (str): The input audio url
             api_key (str, optional): The api api_key. Defaults to None.
             workspace (str): The dashscope workspace id.
             extra_input (Dict): The extra input parameters.
@@ -76,6 +78,7 @@ class VideoSynthesis(BaseAsyncApi):
         return super().call(model,
                             prompt,
                             img_url=img_url,
+                            audio_url=audio_url,
                             api_key=api_key,
                             extend_prompt=extend_prompt,
                             negative_prompt=negative_prompt,
@@ -94,6 +97,7 @@ class VideoSynthesis(BaseAsyncApi):
                    model: str,
                    prompt: Any = None,
                    img_url: str = None,
+                   audio_url: str = None,
                    # """@deprecated, use prompt_extend in parameters """
                    extend_prompt: bool = True,
                    negative_prompt: str = None,
@@ -125,6 +129,13 @@ class VideoSynthesis(BaseAsyncApi):
                 has_upload = True
             inputs['img_url'] = res_img_url
 
+        if audio_url is not None and audio_url:
+            is_upload, res_audio_url = check_and_upload_local(
+                model, audio_url, api_key)
+            if is_upload:
+                has_upload = True
+            inputs['audio_url'] = res_audio_url
+
         if head_frame is not None and head_frame:
             is_upload, res_head_frame = check_and_upload_local(
                 model, head_frame, api_key)
@@ -172,6 +183,7 @@ class VideoSynthesis(BaseAsyncApi):
                    model: str,
                    prompt: Any = None,
                    img_url: str = None,
+                   audio_url: str = None,
                    # """@deprecated, use prompt_extend in parameters """
                    extend_prompt: bool = True,
                    negative_prompt: str = None,
@@ -194,6 +206,7 @@ class VideoSynthesis(BaseAsyncApi):
             negative_prompt (str): The negative prompt is the opposite of the prompt meaning.
             template (str): LoRa input, such as gufeng, katong, etc.
             img_url (str): The input image url, Generate the URL of the image referenced by the video.
+            audio_url (str): The input audio url.
             api_key (str, optional): The api api_key. Defaults to None.
             workspace (str): The dashscope workspace id.
             extra_input (Dict): The extra input parameters.
@@ -215,7 +228,7 @@ class VideoSynthesis(BaseAsyncApi):
         task_group, function = _get_task_group_and_task(__name__)
 
         inputs, kwargs, task = cls._get_input(
-            model, prompt, img_url, extend_prompt, negative_prompt, template, api_key,
+            model, prompt, img_url, audio_url, extend_prompt, negative_prompt, template, api_key,
             extra_input, task, function, head_frame, tail_frame,
             first_frame_url, last_frame_url, **kwargs)
 
@@ -339,6 +352,7 @@ class AioVideoSynthesis(BaseAsyncAioApi):
                    model: str,
                    prompt: Any = None,
                    img_url: str = None,
+                   audio_url: str = None,
                    # """@deprecated, use prompt_extend in parameters """
                    extend_prompt: bool = True,
                    negative_prompt: str = None,
@@ -361,6 +375,7 @@ class AioVideoSynthesis(BaseAsyncAioApi):
             negative_prompt (str): The negative prompt is the opposite of the prompt meaning.
             template (str): LoRa input, such as gufeng, katong, etc.
             img_url (str): The input image url, Generate the URL of the image referenced by the video.
+            audio_url (str): The input audio url.
             api_key (str, optional): The api api_key. Defaults to None.
             workspace (str): The dashscope workspace id.
             extra_input (Dict): The extra input parameters.
@@ -380,7 +395,7 @@ class AioVideoSynthesis(BaseAsyncAioApi):
         """
         task_group, f = _get_task_group_and_task(__name__)
         inputs, kwargs, task = VideoSynthesis._get_input(
-            model, prompt, img_url, extend_prompt, negative_prompt, template, api_key,
+            model, prompt, img_url, audio_url, extend_prompt, negative_prompt, template, api_key,
             extra_input, task, f, head_frame, tail_frame,
             first_frame_url, last_frame_url, **kwargs)
         response = await super().call(model, inputs, task_group, task, f, api_key, workspace, **kwargs)
@@ -391,6 +406,7 @@ class AioVideoSynthesis(BaseAsyncAioApi):
                    model: str,
                    prompt: Any = None,
                    img_url: str = None,
+                   audio_url: str = None,
                    # """@deprecated, use prompt_extend in parameters """
                    extend_prompt: bool = True,
                    negative_prompt: str = None,
@@ -413,6 +429,7 @@ class AioVideoSynthesis(BaseAsyncAioApi):
             negative_prompt (str): The negative prompt is the opposite of the prompt meaning.
             template (str): LoRa input, such as gufeng, katong, etc.
             img_url (str): The input image url, Generate the URL of the image referenced by the video.
+            audio_url (str): The input audio url.
             api_key (str, optional): The api api_key. Defaults to None.
             workspace (str): The dashscope workspace id.
             extra_input (Dict): The extra input parameters.
@@ -434,7 +451,7 @@ class AioVideoSynthesis(BaseAsyncAioApi):
         task_group, function = _get_task_group_and_task(__name__)
 
         inputs, kwargs, task = VideoSynthesis._get_input(
-            model, prompt, img_url, extend_prompt, negative_prompt, template, api_key,
+            model, prompt, img_url, audio_url, extend_prompt, negative_prompt, template, api_key,
             extra_input, task, function, head_frame, tail_frame,
             first_frame_url, last_frame_url, **kwargs)
 

dashscope/app/application.py

@@ -196,4 +196,8 @@ class Application(BaseApi):
         if image_list is not None and image_list:
             input_param['image_list'] = image_list
 
+        file_list = kwargs.pop('file_list', None)
+        if file_list is not None and file_list:
+            input_param['file_list'] = file_list
+
         return input_param, {**parameters, **kwargs}

dashscope/audio/qwen_tts_realtime/qwen_tts_realtime.py

@@ -158,6 +158,7 @@ class QwenTtsRealtime:
                        response_format: AudioFormat = AudioFormat.
                        PCM_24000HZ_MONO_16BIT,
                        mode: str = 'server_commit',
+                       language_type: str = None,
                        **kwargs) -> None:
         '''
         update session configuration, should be used before create response
@@ -170,6 +171,8 @@ class QwenTtsRealtime:
             output audio format
         mode: str
             response mode, server_commit or commit
+        language_type: str
+            language type for synthesized audio, default is 'auto'
         '''
         self.config = {
             'voice': voice,
@@ -177,6 +180,8 @@ class QwenTtsRealtime:
             'response_format': response_format.format,
             'sample_rate': response_format.sample_rate,
         }
+        if language_type is not None:
+            self.config['language_type'] = language_type
         self.config.update(kwargs)
         self.__send_str(
             json.dumps({

dashscope/version.py

@@ -1,3 +1,3 @@
 # Copyright (c) Alibaba, Inc. and its affiliates.
 
-__version__ = '1.24.5'
+__version__ = '1.24.7'

{dashscope-1.24.5.dist-info → dashscope-1.24.7.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dashscope
-Version: 1.24.5
+Version: 1.24.7
 Summary: dashscope client sdk library
 Home-page: https://dashscope.aliyun.com/
 Author: Alibaba Cloud

{dashscope-1.24.5.dist-info → dashscope-1.24.7.dist-info}/RECORD

@@ -3,15 +3,15 @@ dashscope/cli.py,sha256=64oGkevgX0RHPPmMg0sevXDgaFLQNA_0vdtjQ7Z2pHM,26492
 dashscope/files.py,sha256=vRDQygm3lOqBZR73o7KNHs1iTBVuvLncuwJNxIYjzAU,3981
 dashscope/model.py,sha256=B5v_BtYLPqj6raClejBgdKg6WTGwhH_f-20pvsQqmsk,1491
 dashscope/models.py,sha256=dE4mzXkl85G343qVylSGpURPRdA5pZSqXlx6PcxqC_Q,1275
-dashscope/version.py,sha256=2fvqw7bZLyWOIDvUb8DEkdi6y_VgyljhOeYdITEksWM,74
+dashscope/version.py,sha256=bwLuOw_5t9V66aTp9e4fg40jaUHqhx-IXawXHN8THGg,74
 dashscope/aigc/__init__.py,sha256=kYvYEoRK-NUHyMWpBDNQBz4fVA__uOhHRK2kDTBaWgk,617
 dashscope/aigc/chat_completion.py,sha256=ONlyyssIbfaKKcFo7cEKhHx5OCF2XX810HFzIExW1ho,14813
 dashscope/aigc/code_generation.py,sha256=p_mxDKJLQMW0IjFD46JRlZuEZCRESSVKEfLlAevBtqw,10936
 dashscope/aigc/conversation.py,sha256=95xEEY4ThZJysj5zy3aMw7ql9KLJVfD_1iHv9QZ17Ew,14282
 dashscope/aigc/generation.py,sha256=xMcMu16rICTdjZiD_sPqYV_Ltdp4ewGzzfC7JD9VApY,17948
-dashscope/aigc/image_synthesis.py,sha256=Itx9h5brEwC-d3Mj_ntDHGd4qaitqDg9DeGHMJouhMk,28178
-dashscope/aigc/multimodal_conversation.py,sha256=BXpUthyGxJHQs18-m_ZzAw6MI5nSM4_NVMUfTDSC1-k,11682
-dashscope/aigc/video_synthesis.py,sha256=RSPjar5-YiF9xclRmf9H7-5QbRxLcsNXO4zS7oTKi2I,24137
+dashscope/aigc/image_synthesis.py,sha256=sKKKmc11nwNSBEuhLb6ymUr5pxVAAl87uB-ALuh0dLE,28851
+dashscope/aigc/multimodal_conversation.py,sha256=k-OPWwgrWJBb7LiUdFuuP9X6sR5ukbaRHPqFTSikO7Y,12380
+dashscope/aigc/video_synthesis.py,sha256=Wps8ZE3rDRFoebh7w1uvLNSyKHKk1BCTXmq7ryG9iPQ,24885
 dashscope/api_entities/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 dashscope/api_entities/aiohttp_request.py,sha256=1L7XdIJ9L65cQmX8x9JCR4t5hNIMDrbiWADfKKp9yfo,10280
 dashscope/api_entities/api_request_data.py,sha256=04rpYPNK1HkT3iTPJmZpquH621xcBbe8R8EGrDJSLt0,5514
@@ -23,7 +23,7 @@ dashscope/api_entities/encryption.py,sha256=rUCZx3wwVvS5oyKXEeWgyWPxM8Y5d4AaVdgx
 dashscope/api_entities/http_request.py,sha256=MTxYsbkK8oYWDp8ZPjrkdY9YbnQ9SEIy87riyJidMXo,16484
 dashscope/api_entities/websocket_request.py,sha256=PS0FU854-HjTbKa68f4GHa7-noFRMzKySJGfPkrrBjw,16146
 dashscope/app/__init__.py,sha256=xvSvU8O7m5u7vgIvJXTJektJZxmjT2Rpt_YwePH88XE,113
-dashscope/app/application.py,sha256=Whf_ij4RHOaY12_xdS8uj8HVNCwkTp_MRdrFTryF1Kg,9472
+dashscope/app/application.py,sha256=79wnnvAAjWBJ8TtMUN6-B4FHBV7twkAOzhXg7yDJQxE,9620
 dashscope/app/application_response.py,sha256=XO6iOZlt7OXulvFS71zwAq_HXYkn3HLJdAimTWPP0B4,8568
 dashscope/assistants/__init__.py,sha256=hjCTuv13yFaXyUqlexAU-RaO0Ahq3P7VK9_LkSbkGVU,434
 dashscope/assistants/assistant_types.py,sha256=qVDSy0xcsMq_sAD7t_ppoGLBN2QDiHqarAAlW_CDDtY,4478
@@ -41,7 +41,7 @@ dashscope/audio/qwen_omni/omni_realtime.py,sha256=eBmoOxuKcfzMHuXsQWCrIIKmso9iEz
 dashscope/audio/qwen_tts/__init__.py,sha256=JS3axY1grqO0aTIJufZ3KS1JsU6yf6y4K2CQlNvUK9I,132
 dashscope/audio/qwen_tts/speech_synthesizer.py,sha256=7LHR-PXhn-VE1cCOp_82Jq0zE9rMc3xy3dszUeyLLNs,2927
 dashscope/audio/qwen_tts_realtime/__init__.py,sha256=vVkmeJr_mEAn_O0Rh5AU3ICg6qIZqppUryJ5lY8VYPo,254
-dashscope/audio/qwen_tts_realtime/qwen_tts_realtime.py,sha256=8bOAMcDasTHwSLb9xAGJoj9eUPpQHh2aWvonV6Kf1U4,10367
+dashscope/audio/qwen_tts_realtime/qwen_tts_realtime.py,sha256=uMLglxRjUZgol9Z7MTUY1Ji_HvaQaayjYkZvRkrNpkQ,10606
 dashscope/audio/tts/__init__.py,sha256=xYpMFseUZGgqgj_70zcX2VsLv-L7qxJ3d-bbdj_hO0I,245
 dashscope/audio/tts/speech_synthesizer.py,sha256=vD1xQV-rew8qAsIaAGH5amsNtB0SqdtNhVHhJHGQ-xk,7622
 dashscope/audio/tts_v2/__init__.py,sha256=me9a3_7KsHQxcJ8hx4SeKlY1e_ThHVvGMw7Yn0uoscM,333
@@ -100,9 +100,9 @@ dashscope/tokenizers/tokenizer.py,sha256=3FQVDvMNkCW9ccYeJdjrd_PIMMD3Xv7aNZkaYOE
 dashscope/tokenizers/tokenizer_base.py,sha256=5EJIFuizMWESEmLmbd38yJnfeHmPnzZPwsO4aOGjpl4,707
 dashscope/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 dashscope/utils/oss_utils.py,sha256=aZIHlMN2JOfVw6kp0SVrMw_N1MfoTcR_-wiRbJ7DgHw,7501
-dashscope-1.24.5.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-dashscope-1.24.5.dist-info/METADATA,sha256=eRxoK1TphMD4hn-vUM3aPMtaCsvCzstnDe-QPH9A4Q0,7146
-dashscope-1.24.5.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-dashscope-1.24.5.dist-info/entry_points.txt,sha256=e9C3sOf9zDYL0O5ROEGX6FT8w-QK_kaGRWmPZDHAFys,49
-dashscope-1.24.5.dist-info/top_level.txt,sha256=woqavFJK9zas5xTqynmALqOtlafghjsk63Xk86powTU,10
-dashscope-1.24.5.dist-info/RECORD,,
+dashscope-1.24.7.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+dashscope-1.24.7.dist-info/METADATA,sha256=3YnzYbPbSB-pNzn3CNw_Ka5MTDXM8oc_-ddrkugwav4,7146
+dashscope-1.24.7.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+dashscope-1.24.7.dist-info/entry_points.txt,sha256=e9C3sOf9zDYL0O5ROEGX6FT8w-QK_kaGRWmPZDHAFys,49
+dashscope-1.24.7.dist-info/top_level.txt,sha256=woqavFJK9zas5xTqynmALqOtlafghjsk63Xk86powTU,10
+dashscope-1.24.7.dist-info/RECORD,,