feat: Integrate with Tencent SentienceRecognition model

apps/locales/en_US/LC_MESSAGES/django.po

@@ -8570,4 +8570,82 @@ msgid "Edit user authorization status of resource"
 msgstr ""
 
 msgid "Get user authorization status of resource by page"
+msgstr ""
+
+msgid "Obtain resource authorization list by page"
+msgstr ""
+
+msgid "Engine model type"
+msgstr ""
+
+msgid "If not passed, the default value is 16k_zh (Chinese universal)"
+msgstr ""
+
+msgid "Chinese telephone universal"
+msgstr ""
+
+msgid "English telephone universal"
+msgstr ""
+
+msgid "Commonly used in Chinese"
+msgstr ""
+
+msgid "Chinese, English, and Guangdong"
+msgstr ""
+
+msgid "Chinese medical"
+msgstr ""
+
+msgid "English"
+msgstr ""
+
+msgid "Cantonese"
+msgstr ""
+
+msgid "Japanese"
+msgstr ""
+
+msgid "Korean"
+msgstr ""
+
+msgid "Vietnamese"
+msgstr ""
+
+msgid "Malay language"
+msgstr ""
+
+msgid "Indonesian language"
+msgstr ""
+
+msgid "Filipino language"
+msgstr ""
+
+msgid "Thai"
+msgstr ""
+
+msgid "Portuguese"
+msgstr ""
+
+msgid "Turkish"
+msgstr ""
+
+msgid "Arabic"
+msgstr ""
+
+msgid "Spanish"
+msgstr ""
+
+msgid "Hindi"
+msgstr ""
+
+msgid "French"
+msgstr ""
+
+msgid "German"
+msgstr ""
+
+msgid "Multiple dialects, supporting 23 dialects"
+msgstr ""
+
+msgid "This interface is used to recognize short audio files within 60 seconds. Supports Mandarin Chinese, English, Cantonese, Japanese, Vietnamese, Malay, Indonesian, Filipino, Thai, Portuguese, Turkish, Arabic, Hindi, French, German, and 23 Chinese dialects."
 msgstr ""

apps/locales/zh_CN/LC_MESSAGES/django.po

@@ -8696,4 +8696,82 @@ msgid "Edit user authorization status of resource"
 msgstr "修改资源对用户的授权状态"
 
 msgid "Get user authorization status of resource by page"
-msgstr "分页获取资源对用户的授权状态"
+msgstr "分页获取资源对用户的授权状态"
+
+msgid "Obtain resource authorization list by page"
+msgstr "分页获取资源授权列表"
+
+msgid "Engine model type"
+msgstr "引擎模型类型"
+
+msgid "If not passed, the default value is 16k_zh (Chinese universal)"
+msgstr "如果未传递，默认值为 16k_zh（中文通用）"
+
+msgid "Chinese telephone universal"
+msgstr "中文电话通用"
+
+msgid "English telephone universal"
+msgstr "英文电话通用"
+
+msgid "Commonly used in Chinese"
+msgstr "中文常用"
+
+msgid "Chinese, English, and Guangdong"
+msgstr "中文、英文和广东话"
+
+msgid "Chinese medical"
+msgstr "中文医疗"
+
+msgid "English"
+msgstr "英文"
+
+msgid "Cantonese"
+msgstr "粤语"
+
+msgid "Japanese"
+msgstr "日语"
+
+msgid "Korean"
+msgstr "韩语"
+
+msgid "Vietnamese"
+msgstr "越南语"
+
+msgid "Malay language"
+msgstr "马来语"
+
+msgid "Indonesian language"
+msgstr "印尼语"
+
+msgid "Filipino language"
+msgstr "菲律宾语"
+
+msgid "Thai"
+msgstr "泰语"
+
+msgid "Portuguese"
+msgstr "葡萄牙语"
+
+msgid "Turkish"
+msgstr "土耳其语"
+
+msgid "Arabic"
+msgstr "阿拉伯语"
+
+msgid "Spanish"
+msgstr "西班牙语"
+
+msgid "Hindi"
+msgstr "印地语"
+
+msgid "French"
+msgstr "法语"
+
+msgid "German"
+msgstr "德语"
+
+msgid "Multiple dialects, supporting 23 dialects"
+msgstr "多种方言，支持 23 种方言"
+
+msgid "This interface is used to recognize short audio files within 60 seconds. Supports Mandarin Chinese, English, Cantonese, Japanese, Vietnamese, Malay, Indonesian, Filipino, Thai, Portuguese, Turkish, Arabic, Hindi, French, German, and 23 Chinese dialects."
+msgstr "本接口用于识别 60 秒之内的短音频文件。支持中文普通话、英语、粤语、日语、越南语、马来语、印度尼西亚语、菲律宾语、泰语、葡萄牙语、土耳其语、阿拉伯语、印地语、法语、德语及 23 种汉语方言。"

apps/locales/zh_Hant/LC_MESSAGES/django.po

@@ -8696,4 +8696,82 @@ msgid "Edit user authorization status of resource"
 msgstr "修改資源對用戶的授權狀態"
 
 msgid "Get user authorization status of resource by page"
-msgstr "分頁獲取資源對用戶的授權狀態"
+msgstr "分頁獲取資源對用戶的授權狀態"
+
+msgid "Obtain resource authorization list by page"
+msgstr "分頁獲取資源授權清單"
+
+msgid "Engine model type"
+msgstr "引擎模型類型"
+
+msgid "If not passed, the default value is 16k_zh (Chinese universal)"
+msgstr "如果未傳遞，默認值為 16k_zh（中文通用）"
+
+msgid "Chinese telephone universal"
+msgstr "中文電話通用"
+
+msgid "English telephone universal"
+msgstr "英文電話通用"
+
+msgid "Commonly used in Chinese"
+msgstr "中文常用"
+
+msgid "Chinese, English, and Guangdong"
+msgstr "中文、英文和廣東話"
+
+msgid "Chinese medical"
+msgstr "中文醫療"
+
+msgid "English"
+msgstr "英文"
+
+msgid "Cantonese"
+msgstr "粵語"
+
+msgid "Japanese"
+msgstr "日語"
+
+msgid "Korean"
+msgstr "韓語"
+
+msgid "Vietnamese"
+msgstr "越南語"
+
+msgid "Malay language"
+msgstr "馬來語"
+
+msgid "Indonesian language"
+msgstr "印尼語"
+
+msgid "Filipino language"
+msgstr "菲律賓語"
+
+msgid "Thai"
+msgstr "泰語"
+
+msgid "Portuguese"
+msgstr "葡萄牙語"
+
+msgid "Turkish"
+msgstr "土耳其語"
+
+msgid "Arabic"
+msgstr "阿拉伯語"
+
+msgid "Spanish"
+msgstr "西班牙語"
+
+msgid "Hindi"
+msgstr "印地語"
+
+msgid "French"
+msgstr "法語"
+
+msgid "German"
+msgstr "德語"
+
+msgid "Multiple dialects, supporting 23 dialects"
+msgstr "多種方言，支持 23 種方言"
+
+msgid "This interface is used to recognize short audio files within 60 seconds. Supports Mandarin Chinese, English, Cantonese, Japanese, Vietnamese, Malay, Indonesian, Filipino, Thai, Portuguese, Turkish, Arabic, Hindi, French, German, and 23 Chinese dialects."
+msgstr "本介面用於識別 60 秒之內的短音頻文件。支援中文普通話、英語、粵語、日語、越南語、馬來語、印度尼西亞語、菲律賓語、泰語、葡萄牙語、土耳其語、阿拉伯語、印地語、法語、德語及 23 種漢語方言。"

apps/models_provider/impl/tencent_model_provider/credential/stt.py

@@ -0,0 +1,90 @@
+import traceback
+
+from common import forms
+from common.exception.app_exception import AppApiException
+from common.forms import BaseForm, TooltipLabel
+from django.utils.translation import gettext_lazy as _, gettext
+
+from models_provider.base_model_provider import BaseModelCredential, ValidCode
+
+
+class TencentSSTModelParams(BaseForm):
+    EngSerViceType = forms.SingleSelect(
+        TooltipLabel(_('Engine model type'), _('If not passed, the default value is 16k_zh (Chinese universal)')),
+        required=True,
+        default_value='16k_zh',
+        option_list=[
+            {"value": "8k_zh", "label": _("Chinese telephone universal")},
+            {"value": "8k_en", "label": _("English telephone universal")},
+            {"value": "16k_zh", "label": _("Commonly used in Chinese")},
+            {"value": "16k_zh-PY", "label": _("Chinese, English, and Guangdong")},
+            {"value": "16k_zh_medical", "label": _("Chinese medical")},
+            {"value": "16k_en", "label": _("English")},
+            {"value": "16k_yue", "label": _("Cantonese")},
+            {"value": "16k_ja", "label": _("Japanese")},
+            {"value": "16k_ko", "label": _("Korean")},
+            {"value": "16k_vi", "label": _("Vietnamese")},
+            {"value": "16k_ms", "label": _("Malay language")},
+            {"value": "16k_id", "label": _("Indonesian language")},
+            {"value": "16k_fil", "label": _("Filipino language")},
+            {"value": "16k_th", "label": _("Thai")},
+            {"value": "16k_pt", "label": _("Portuguese")},
+            {"value": "16k_tr", "label": _("Turkish")},
+            {"value": "16k_ar", "label": _("Arabic")},
+            {"value": "16k_es", "label": _("Spanish")},
+            {"value": "16k_hi", "label": _("Hindi")},
+            {"value": "16k_fr", "label": _("French")},
+            {"value": "16k_de", "label": _("German")},
+            {"value": "16k_zh_dialect", "label": _("Multiple dialects, supporting 23 dialects")}
+        ],
+        value_field='value',
+        text_field='label'
+    )
+
+class TencentSTTModelCredential(BaseForm, BaseModelCredential):
+    REQUIRED_FIELDS = ["SecretId", "SecretKey"]
+
+    @classmethod
+    def _validate_model_type(cls, model_type, provider, raise_exception=False):
+        if not any(mt['value'] == model_type for mt in provider.get_model_type_list()):
+            if raise_exception:
+                raise AppApiException(ValidCode.valid_error.value,
+                                      gettext('{model_type} Model type is not supported').format(model_type=model_type))
+            return False
+        return True
+
+    @classmethod
+    def _validate_credential_fields(cls, model_credential, raise_exception=False):
+        missing_keys = [key for key in cls.REQUIRED_FIELDS if key not in model_credential]
+        if missing_keys:
+            if raise_exception:
+                raise AppApiException(ValidCode.valid_error.value,
+                                      gettext('{keys} is required').format(keys=", ".join(missing_keys)))
+            return False
+        return True
+
+    def is_valid(self, model_type, model_name, model_credential, model_params, provider, raise_exception=False):
+        if not (self._validate_model_type(model_type, provider, raise_exception) and
+                self._validate_credential_fields(model_credential, raise_exception)):
+            return False
+        try:
+            model = provider.get_model(model_type, model_name, model_credential, **model_params)
+            model.check_auth()
+        except Exception as e:
+            traceback.print_exc()
+            if raise_exception:
+                raise AppApiException(ValidCode.valid_error.value,
+                                      gettext(
+                                          'Verification failed, please check whether the parameters are correct: {error}').format(
+                                          error=str(e)))
+            return False
+        return True
+
+    def encryption_dict(self, model):
+        return {**model, 'SecretKey': super().encryption(model.get('SecretKey', ''))}
+
+    SecretId = forms.PasswordInputField('SecretId', required=True)
+    SecretKey = forms.PasswordInputField('SecretKey', required=True)
+
+    def get_model_params_setting_form(self, model_name):
+        return TencentSSTModelParams()

apps/models_provider/impl/tencent_model_provider/model/stt.py

@@ -0,0 +1,80 @@
+import base64
+import json
+import os
+from typing import Dict
+
+from tencentcloud.asr.v20190614 import asr_client, models
+from tencentcloud.common import credential
+from tencentcloud.common.exception import TencentCloudSDKException
+from tencentcloud.common.profile.client_profile import ClientProfile
+from tencentcloud.common.profile.http_profile import HttpProfile
+
+from models_provider.base_model_provider import MaxKBBaseModel
+from models_provider.impl.base_stt import BaseSpeechToText
+
+
+class TencentSpeechToText(MaxKBBaseModel, BaseSpeechToText):
+    hunyuan_secret_id: str
+    hunyuan_secret_key: str
+    model: str
+    params: dict
+
+    def __init__(self, **kwargs):
+        super().__init__(**kwargs)
+        self.hunyuan_secret_id = kwargs.get('hunyuan_secret_id')
+        self.hunyuan_secret_key = kwargs.get('hunyuan_secret_key')
+        self.model = kwargs.get('model')
+        self.params = kwargs.get('params')
+
+    @staticmethod
+    def is_cache_model():
+        return False
+
+    @staticmethod
+    def new_instance(model_type, model_name, model_credential: Dict[str, object], **model_kwargs):
+        return TencentSpeechToText(
+            hunyuan_secret_id=model_credential.get('SecretId'),
+            hunyuan_secret_key=model_credential.get('SecretKey'),
+            model=model_name,
+            params=model_kwargs,
+            **model_kwargs
+        )
+
+    def check_auth(self):
+        cwd = os.path.dirname(os.path.abspath(__file__))
+        with open(f'{cwd}/iat_mp3_16k.mp3', 'rb') as f:
+            self.speech_to_text(f)
+
+    def speech_to_text(self, audio_file):
+        try:
+            cred = credential.Credential(self.hunyuan_secret_id, self.hunyuan_secret_key)
+            # 实例化一个http选项，可选的，没有特殊需求可以跳过
+            httpProfile = HttpProfile()
+            httpProfile.endpoint = "asr.tencentcloudapi.com"
+
+            # 实例化一个client选项，可选的，没有特殊需求可以跳过
+            clientProfile = ClientProfile()
+            clientProfile.httpProfile = httpProfile
+            # 实例化要请求产品的client对象,clientProfile是可选的
+            client = asr_client.AsrClient(cred, "", clientProfile)
+            buf = audio_file.read()
+            _v = base64.b64encode(buf)
+
+            # 实例化一个请求对象,每个接口都会对应一个request对象
+            req = models.SentenceRecognitionRequest()
+            params = {
+                "EngSerViceType": self.params.get('EngSerViceType'),
+                "SourceType": 1,
+                "VoiceFormat": "mp3",
+                "Data": _v.decode(),
+            }
+            req.from_json_string(json.dumps(params))
+
+            # 返回的resp是一个SentenceRecognitionResponse的实例，与请求对象对应
+            resp = client.SentenceRecognition(req)
+            # 输出json格式的字符串回包
+            return resp.Result
+
+
+        except TencentCloudSDKException as err:
+            print(err)

apps/models_provider/impl/tencent_model_provider/tencent_model_provider.py

@@ -9,10 +9,12 @@ from models_provider.base_model_provider import (
 from models_provider.impl.tencent_model_provider.credential.embedding import TencentEmbeddingCredential
 from models_provider.impl.tencent_model_provider.credential.image import TencentVisionModelCredential
 from models_provider.impl.tencent_model_provider.credential.llm import TencentLLMModelCredential
+from models_provider.impl.tencent_model_provider.credential.stt import TencentSTTModelCredential
 from models_provider.impl.tencent_model_provider.credential.tti import TencentTTIModelCredential
 from models_provider.impl.tencent_model_provider.model.embedding import TencentEmbeddingModel
 from models_provider.impl.tencent_model_provider.model.image import TencentVision
 from models_provider.impl.tencent_model_provider.model.llm import TencentModel
+from models_provider.impl.tencent_model_provider.model.stt import TencentSpeechToText
 from models_provider.impl.tencent_model_provider.model.tti import TencentTextToImageModel
 from maxkb.conf import PROJECT_DIR
 from django.utils.translation import gettext as _
@@ -70,6 +72,12 @@ def _initialize_model_info():
             ModelTypeConst.LLM,
             TencentLLMModelCredential,
             TencentModel),
+        _create_model_info(
+            'asr-sentence',
+            _("This interface is used to recognize short audio files within 60 seconds. Supports Mandarin Chinese, English, Cantonese, Japanese, Vietnamese, Malay, Indonesian, Filipino, Thai, Portuguese, Turkish, Arabic, Hindi, French, German, and 23 Chinese dialects."),
+            ModelTypeConst.STT,
+            TencentSTTModelCredential,
+            TencentSpeechToText),
     ]
 
     tencent_embedding_model_info = _create_model_info(

apps/models_provider/tools.py

@@ -107,7 +107,7 @@ def is_valid_credential(provider, model_type, model_name, model_credential: Dict
 def get_model_by_id(_id, workspace_id):
     model = QuerySet(Model).filter(id=_id).first()
     get_authorized_model = DatabaseModelManage.get_model("get_authorized_model")
-    if model and model.workspace_id!=workspace_id and get_authorized_model is not None:
+    if model and model.workspace_id != workspace_id and get_authorized_model is not None:
         model = get_authorized_model(QuerySet(Model).filter(id=_id), workspace_id).first()
     if model is None:
         raise Exception(_("Model does not exist"))
@@ -122,4 +122,5 @@ def get_model_instance_by_model_workspace_id(model_id, workspace_id, **kwargs):
     @return:                模型实例
     """
     model = get_model_by_id(model_id, workspace_id)
-    return ModelManage.get_model(model_id, lambda _id: get_model(model, **kwargs))
+    s = {p.get('field'): p.get('default_value') for p in model.model_params_form if p.get('default_value') is not None}
+    return ModelManage.get_model(model_id, lambda _id: get_model(model, **{**s, **kwargs}))

pyproject.toml

@@ -76,6 +76,7 @@ websockets = "15.0.1"
 
 # 开发工具
 pylint = "3.3.7"
+tencentcloud-sdk-python-asr = "^3.0.1416"
 
 
 [build-system]