diff --git a/dsLightRag/KeDaXunFei/XunFeiAudioEvaluator_cn.py b/dsLightRag/KeDaXunFei/XunFeiAudioEvaluator_cn.py index 9378b946..cbf479e4 100644 --- a/dsLightRag/KeDaXunFei/XunFeiAudioEvaluator_cn.py +++ b/dsLightRag/KeDaXunFei/XunFeiAudioEvaluator_cn.py @@ -17,18 +17,16 @@ from Config.Config import XF_APPID, XF_APISECRET, XF_APIKEY class XunFeiAudioEvaluator_cn: - """讯飞语音评测类""" - - def __init__(self, appid, api_key, api_secret, audio_file, language, txt): + """讯飞语音评测类（中文专用）""" + def __init__(self, appid, api_key, api_secret, audio_file, txt): self.appid = appid self.api_key = api_key self.api_secret = api_secret self.audio_file = audio_file - self.language = language + self.txt = txt.strip() # 确保文本预处理 self.host_url = "wss://ise-api.xfyun.cn/v2/open-ise" + self.evaluation_results = {} # 统一结果存储变量 self.websocket_url = "" - self.evaluation_results = {} - self.txt = txt def generate_auth_url(self): """生成鉴权URL""" @@ -98,10 +96,10 @@ class XunFeiAudioEvaluator_cn: print(f"WebSocket connection closed,{ws}") def on_open(self, ws): - """连接建立处理""" + """连接建立处理（中文专用配置）""" print(f"WebSocket connection opened,{ws},ws连接建立成功...") - # 发送初始参数 + # 发送初始参数（中文专用配置） send_dict = { "common": { "app_id": self.appid @@ -110,13 +108,14 @@ class XunFeiAudioEvaluator_cn: "category": "read_sentence", "rstcd": "utf8", "sub": "ise", - "group": "pupil", - "ent": "cn_vip" if self.language == "chinese" else "en_vip", + "group": "pupil", # 根据文档设置为小学组 + "ent": "cn_vip", # 固定中文引擎 "tte": "utf-8", "cmd": "ssb", - "auf": "audio/L16;rate=16000", - "aue": "lame", - "text": '\uFEFF' + "[content]\n" + self.txt + "auf": "audio/L16;rate=16000", # 修复参数格式错误 + "aue": "raw", # 中文推荐使用raw格式 + "text": '\uFEFF' + "[content]\n" + self.txt, # 确保UTF8 BOM + "extra_ability": "multi_dimension" # 启用多维度评分 }, "data": { "status": 0, @@ -167,21 +166,38 @@ class XunFeiAudioEvaluator_cn: time.sleep(0.04) def parse_evaluation_results(self, xml_content): - root = ET.fromstring(xml_content) - # 兼容中英文不同的根节点 - read_node = root.find('.//read_sentence') or root.find('.//read_chapter') - if read_node is not None: - self.results = { - 'total_score': read_node.attrib.get('total_score', '0'), - 'accuracy': read_node.attrib.get('accuracy_score', '0'), - 'fluency': read_node.attrib.get('fluency_score', '0'), - 'integrity': read_node.attrib.get('integrity_score', '0') - } - else: - # 打印未匹配的XML结构用于调试 - print(f"No evaluation nodes found in XML: {xml_content}") - self.results = {'error': '未找到有效评分节点'} - """解析评测结果XML并提取得分信息""" + """解析中文评测XML结果（仅处理read_sentence节点）""" + try: + root = ET.fromstring(xml_content) + # 中文专用：仅查找read_sentence节点 + read_sentence = root.find('.//read_sentence') + if read_sentence is not None: + # 中文评分字段映射 + self.evaluation_results = { + 'total_score': float(read_sentence.get('total_score', 0)), + 'accuracy_score': float(read_sentence.get('accuracy_score', 0)), + 'fluency_score': float(read_sentence.get('fluency_score', 0)), + 'integrity_score': float(read_sentence.get('integrity_score', 0)), + 'tone_score': float(read_sentence.get('tone_score', 0)), # 中文特有调型分 + 'is_rejected': read_sentence.get('is_rejected', 'false') == 'true' + } + # 提取句子级得分（中文特有） + sentences = [] + for sent in read_sentence.findall('.//sentence'): + sentences.append({ + 'content': sent.get('content', ''), + 'total_score': float(sent.get('total_score', 0)), + 'fluency_score': float(sent.get('fluency_score', 0)) + }) + self.evaluation_results['sentences'] = sentences + print(f"中文评测结果解析成功: {self.evaluation_results}") + else: + print(f"未找到中文评分节点: {xml_content[:200]}") # 打印前200字符调试 + self.evaluation_results = {'error': '未找到read_sentence节点'} + except ET.ParseError as e: + print(f"XML解析失败: {e}, 内容: {xml_content[:200]}") + self.evaluation_results = {'error': f'解析失败: {str(e)}'} + """解析评测结果XML并提得分信息""" try: print(xml_content) root = ET.fromstring(xml_content) @@ -266,7 +282,7 @@ class XunFeiAudioEvaluator_cn: return descriptions.get(dp_message, f"未知({dp_message})") def run_evaluation(self): - """运行评测""" + """运行评测（修复SSL配置）""" start_time = datetime.now() websocket.enableTrace(False) @@ -279,7 +295,12 @@ class XunFeiAudioEvaluator_cn: on_open=self.on_open ) - ws_entity.run_forever(sslopt={"cert_reqs": ssl.CERT_NONE}) + # 修复SSL配置，启用证书验证 + import certifi + ws_entity.run_forever(sslopt={ + "cert_reqs": ssl.CERT_REQUIRED, + "ca_certs": certifi.where() + }) end_time = datetime.now() evaluation_time = end_time - start_time @@ -290,15 +311,15 @@ class XunFeiAudioEvaluator_cn: # 使用示例 if __name__ == '__main__': - # 配置参数 - # appid = "5b83f8d6" - # api_secret = "604fa6cb9c5ab664a0d153fe0ccc6802" - # api_key = "5beb887923204000bfcb402046bb05a6" + # 中文使用示例 appid = XF_APPID api_secret = XF_APISECRET api_key = XF_APIKEY - # audio_file = "./1.mp3" - audio_file = r'D:\dsWork\dsProject\dsLightRag\static\audio\audio_afc0a96e382c428cba2f00e3f71e4e8f.mp3' + audio_file = r'D:\dsWork\dsProject\dsLightRag\static\audio\test_cn.wav' # 中文测试音频 + txt = "大家好。很高兴认识你们。今天天气很好。我正在使用这个工具练习中文发音。" + evaluator = XunFeiAudioEvaluator_cn(appid, api_key, api_secret, audio_file, txt) + results, eval_time = evaluator.run_evaluation() + print(evaluator.get_evaluation_summary()) # 创建评测器实例 txt="Hello everyone! Nice to meet you. Today is a beautiful day. I am learning English pronunciation with this tool." evaluator = XunFeiAudioEvaluator_cn(appid, api_key, api_secret, audio_file, "english", txt) diff --git a/dsLightRag/KeDaXunFei/__pycache__/XunFeiAudioEvaluator_cn.cpython-310.pyc b/dsLightRag/KeDaXunFei/__pycache__/XunFeiAudioEvaluator_cn.cpython-310.pyc new file mode 100644 index 00000000..05ebbcb6 Binary files /dev/null and b/dsLightRag/KeDaXunFei/__pycache__/XunFeiAudioEvaluator_cn.cpython-310.pyc differ diff --git a/dsLightRag/Routes/XunFeiRoute.py b/dsLightRag/Routes/XunFeiRoute.py new file mode 100644 index 00000000..e69de29b diff --git a/dsLightRag/Routes/XunFeiRoute_cn.py b/dsLightRag/Routes/XunFeiRoute_cn.py index 1e326487..cd5b065a 100644 --- a/dsLightRag/Routes/XunFeiRoute_cn.py +++ b/dsLightRag/Routes/XunFeiRoute_cn.py @@ -19,7 +19,7 @@ sys.path.append(os.path.join(os.path.dirname(__file__), "..")) from Config.Config import XF_APPID, XF_APIKEY, XF_APISECRET @router.post("/save-audio") -async def save_audio(audio: UploadFile = File(...), txt: str = Form(...), language: str = Form("english")): # 添加语言参数，默认英文 +async def save_audio(audio: UploadFile = File(...), txt: str = Form(...)): """保存音频文件并评分""" temp_file = None try: @@ -39,16 +39,18 @@ async def save_audio(audio: UploadFile = File(...), txt: str = Form(...), langua # 添加wav转mp3功能 import subprocess - mp3_temp_file = os.path.join(temp_dir, f"temp_{uuid.uuid4().hex}.mp3") + # 修改MP3保存路径为UPLOAD_DIR，使用相同文件名前缀 + mp3_file_name = os.path.splitext(file_name)[0] + ".mp3" + mp3_file_path = os.path.join(UPLOAD_DIR, mp3_file_name) try: # 使用ffmpeg将wav转换为mp3 subprocess.run( - ["ffmpeg", "-i", temp_file, "-y", mp3_temp_file], + ["ffmpeg", "-i", temp_file, "-y", mp3_file_path], check=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE ) - logger.info(f"已将wav转换为mp3: {mp3_temp_file}") + logger.info(f"已将wav转换为mp3: {mp3_file_path}") except subprocess.CalledProcessError as e: logger.error(f"ffmpeg转换失败: {e.stderr.decode()}") raise Exception(f"音频格式转换失败: {e.stderr.decode()}") @@ -59,8 +61,7 @@ async def save_audio(audio: UploadFile = File(...), txt: str = Form(...), langua appid=XF_APPID, api_key=XF_APIKEY, api_secret=XF_APISECRET, - audio_file=mp3_temp_file, - language=language, # 使用动态参数 + audio_file=mp3_file_path, # 使用新的MP3路径 txt=txt ) results, eval_time = evaluator.run_evaluation() diff --git a/dsLightRag/Routes/__pycache__/XunFeiRoute_cn.cpython-310.pyc b/dsLightRag/Routes/__pycache__/XunFeiRoute_cn.cpython-310.pyc new file mode 100644 index 00000000..97d1a6c0 Binary files /dev/null and b/dsLightRag/Routes/__pycache__/XunFeiRoute_cn.cpython-310.pyc differ diff --git a/dsLightRag/Routes/__pycache__/XunFeiRoute_en.cpython-310.pyc b/dsLightRag/Routes/__pycache__/XunFeiRoute_en.cpython-310.pyc new file mode 100644 index 00000000..23c8d20b Binary files /dev/null and b/dsLightRag/Routes/__pycache__/XunFeiRoute_en.cpython-310.pyc differ diff --git a/dsLightRag/static/XunFei/audio_evaluation_cn.html b/dsLightRag/static/XunFei/audio_evaluation_cn.html index 806d863c..423c4d77 100644 --- a/dsLightRag/static/XunFei/audio_evaluation_cn.html +++ b/dsLightRag/static/XunFei/audio_evaluation_cn.html @@ -360,12 +360,11 @@ const fileName = `recording_${Date.now()}.webm`; formData.append('audio', this.recordedBlob, fileName); formData.append('txt', document.getElementById('readingText').value); - formData.append('language', 'chinese'); // 添加中文语言参数 try { this.simulateUploadProgress(); - const response = await fetch('/api/xunFei/save-audio', { + const response = await fetch('/api/xunFeiCn/save-audio', { method: 'POST', body: formData, headers: {'Accept': 'application/json'} diff --git a/dsLightRag/static/audio/audio_00008a1143e64d75bd51e7b0c9e7ce2b.mp3 b/dsLightRag/static/audio/audio_00008a1143e64d75bd51e7b0c9e7ce2b.mp3 new file mode 100644 index 00000000..c7bdc98c Binary files /dev/null and b/dsLightRag/static/audio/audio_00008a1143e64d75bd51e7b0c9e7ce2b.mp3 differ diff --git a/dsLightRag/static/audio/audio_00008a1143e64d75bd51e7b0c9e7ce2b.wav b/dsLightRag/static/audio/audio_00008a1143e64d75bd51e7b0c9e7ce2b.wav new file mode 100644 index 00000000..09b823d8 Binary files /dev/null and b/dsLightRag/static/audio/audio_00008a1143e64d75bd51e7b0c9e7ce2b.wav differ diff --git a/dsLightRag/static/audio/audio_1b1c1b1f92584b55a5ffd2146989dda3.wav b/dsLightRag/static/audio/audio_1b1c1b1f92584b55a5ffd2146989dda3.wav deleted file mode 100644 index cac384ff..00000000 Binary files a/dsLightRag/static/audio/audio_1b1c1b1f92584b55a5ffd2146989dda3.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_1b2f55ad68294cd7b7ad92855dc472a3.wav b/dsLightRag/static/audio/audio_1b2f55ad68294cd7b7ad92855dc472a3.wav deleted file mode 100644 index 8ba56799..00000000 Binary files a/dsLightRag/static/audio/audio_1b2f55ad68294cd7b7ad92855dc472a3.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_1d892f27addd473c9da65f7d6135b349.wav b/dsLightRag/static/audio/audio_1d892f27addd473c9da65f7d6135b349.wav deleted file mode 100644 index 1958a926..00000000 Binary files a/dsLightRag/static/audio/audio_1d892f27addd473c9da65f7d6135b349.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_1e213f49b6f84f96a3adc8eaaac14d61.wav b/dsLightRag/static/audio/audio_1e213f49b6f84f96a3adc8eaaac14d61.wav deleted file mode 100644 index 38e9d5e5..00000000 Binary files a/dsLightRag/static/audio/audio_1e213f49b6f84f96a3adc8eaaac14d61.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_20a9bf39f251448bbce03c28234add3d.wav b/dsLightRag/static/audio/audio_20a9bf39f251448bbce03c28234add3d.wav deleted file mode 100644 index 326b41de..00000000 Binary files a/dsLightRag/static/audio/audio_20a9bf39f251448bbce03c28234add3d.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_2df2cb8caf0d46ccaf3be5f5b37a26ce.wav b/dsLightRag/static/audio/audio_2df2cb8caf0d46ccaf3be5f5b37a26ce.wav deleted file mode 100644 index c158dbea..00000000 Binary files a/dsLightRag/static/audio/audio_2df2cb8caf0d46ccaf3be5f5b37a26ce.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_8986c1cb60234f7692a3fbfacdb920e6.wav b/dsLightRag/static/audio/audio_8986c1cb60234f7692a3fbfacdb920e6.wav deleted file mode 100644 index b5284c32..00000000 Binary files a/dsLightRag/static/audio/audio_8986c1cb60234f7692a3fbfacdb920e6.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_8e7bdfb77bbb4acdb4ae385a195a40f4.wav b/dsLightRag/static/audio/audio_8e7bdfb77bbb4acdb4ae385a195a40f4.wav deleted file mode 100644 index e5c72a4a..00000000 Binary files a/dsLightRag/static/audio/audio_8e7bdfb77bbb4acdb4ae385a195a40f4.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_929659d30cb54668b76185956a0481eb.wav b/dsLightRag/static/audio/audio_929659d30cb54668b76185956a0481eb.wav deleted file mode 100644 index a5f092bc..00000000 Binary files a/dsLightRag/static/audio/audio_929659d30cb54668b76185956a0481eb.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_9d87bc5df79847608cb985e3ed119330.wav b/dsLightRag/static/audio/audio_9d87bc5df79847608cb985e3ed119330.wav deleted file mode 100644 index 267e651d..00000000 Binary files a/dsLightRag/static/audio/audio_9d87bc5df79847608cb985e3ed119330.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_a68241bb23ca4d10bc66c768022ee8a9.wav b/dsLightRag/static/audio/audio_a68241bb23ca4d10bc66c768022ee8a9.wav deleted file mode 100644 index cf6ccd87..00000000 Binary files a/dsLightRag/static/audio/audio_a68241bb23ca4d10bc66c768022ee8a9.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_afc0a96e382c428cba2f00e3f71e4e8f.mp3 b/dsLightRag/static/audio/audio_afc0a96e382c428cba2f00e3f71e4e8f.mp3 deleted file mode 100644 index dbc75607..00000000 Binary files a/dsLightRag/static/audio/audio_afc0a96e382c428cba2f00e3f71e4e8f.mp3 and /dev/null differ diff --git a/dsLightRag/static/audio/audio_afc0a96e382c428cba2f00e3f71e4e8f.wav b/dsLightRag/static/audio/audio_afc0a96e382c428cba2f00e3f71e4e8f.wav deleted file mode 100644 index 31ccb18a..00000000 Binary files a/dsLightRag/static/audio/audio_afc0a96e382c428cba2f00e3f71e4e8f.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_b6a8bad4000c4518873e53714420d78e.wav b/dsLightRag/static/audio/audio_b6a8bad4000c4518873e53714420d78e.wav deleted file mode 100644 index 5f36ed3e..00000000 Binary files a/dsLightRag/static/audio/audio_b6a8bad4000c4518873e53714420d78e.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_d2169f26cad44f7097d4a761dbc26325.wav b/dsLightRag/static/audio/audio_d2169f26cad44f7097d4a761dbc26325.wav deleted file mode 100644 index 4f4851ad..00000000 Binary files a/dsLightRag/static/audio/audio_d2169f26cad44f7097d4a761dbc26325.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_e6905469ec344cc1adb4555c1b90f3cd.wav b/dsLightRag/static/audio/audio_e6905469ec344cc1adb4555c1b90f3cd.wav deleted file mode 100644 index c2f0a731..00000000 Binary files a/dsLightRag/static/audio/audio_e6905469ec344cc1adb4555c1b90f3cd.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_e72bf2fb69fa47fd865b4d5030c2ce94.wav b/dsLightRag/static/audio/audio_e72bf2fb69fa47fd865b4d5030c2ce94.wav new file mode 100644 index 00000000..2aca2606 Binary files /dev/null and b/dsLightRag/static/audio/audio_e72bf2fb69fa47fd865b4d5030c2ce94.wav differ diff --git a/dsLightRag/static/audio/audio_ed5b6a719790471cbd7856edd390142c.wav b/dsLightRag/static/audio/audio_ed5b6a719790471cbd7856edd390142c.wav deleted file mode 100644 index 94eaeca3..00000000 Binary files a/dsLightRag/static/audio/audio_ed5b6a719790471cbd7856edd390142c.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_f32a090081d343c19b15f65f3d5841d1.wav b/dsLightRag/static/audio/audio_f32a090081d343c19b15f65f3d5841d1.wav deleted file mode 100644 index 3209b7c4..00000000 Binary files a/dsLightRag/static/audio/audio_f32a090081d343c19b15f65f3d5841d1.wav and /dev/null differ diff --git a/dsLightRag/static/audio/audio_f69d7ee2fe324d2486d1db61b9128dd2.wav b/dsLightRag/static/audio/audio_f69d7ee2fe324d2486d1db61b9128dd2.wav deleted file mode 100644 index 2486ca01..00000000 Binary files a/dsLightRag/static/audio/audio_f69d7ee2fe324d2486d1db61b9128dd2.wav and /dev/null differ