4 Pull requests merged by 3 people

• feat: support minimax t2a

  #96 merged Jun 11, 2025

• refactor(websocket): extract websocket message types

  #93 merged Jun 9, 2025

• 1：增加ffmpeg安装检查，给出告警提醒

  #92 merged Jun 5, 2025

• 新增讯飞云的语音识别和语音合成

  #89 merged Jun 5, 2025

16 Issues closed by 2 people

• Readme说支持MQTT, 但是代码中没有看到相关实现?

  #97 closed Jun 11, 2025

• 从Coze API获取智能体列表存在userId为空问题

  #95 closed Jun 9, 2025

• 怎么不能实时打断呀

  #91 closed Jun 5, 2025

• Maven编译报错

  #90 closed Jun 5, 2025

• 需对接dify,讯飞，腾讯，等多家llm,tts,stt服务商

  #34 closed Jun 4, 2025

• 用win的客户端py-xiaozhi替换硬件，语音识别乱码，vosk

  #78 closed Jun 2, 2025

• Vosk 加载，java程序总是被杀死

  #88 closed Jun 2, 2025

• 有杂音

  #85 closed May 25, 2025

• 提Bug，用zhipu的模型的时候，应该传apiSecret

  #84 closed May 25, 2025

• X86的Mac怎么办，

  #82 closed May 22, 2025

• OTA升级有详细一些的说明吗?实际测试能请求到但是硬件设备没反应 一直提示检查新版本...

  #80 closed May 20, 2025

• 对话要求调低音量,设备回复不支持,需要哪里设置

  #79 closed May 19, 2025

• 音乐播放功能

  #74 closed May 15, 2025

• 接收到的用户的音频文件路径并没有保存到数据库中

  #40 closed May 14, 2025

• 双向流式对话

  #48 closed May 12, 2025

• 项目运行不起来，会卡主然后直接踢掉线

  #77 closed May 12, 2025

4 Issues opened by 1 person

• 音乐播放时，如果设备直接重启，服务端会一直报发送失败，需要在服务端发送失败后关闭所有任务

  #94 opened Jun 6, 2025

• 语音合成高级设置，同模型超参，可以修改说话速率，pitch等

  #87 opened May 25, 2025

• 模型超参配置功能，可以在配置模型时以高级功能展开的方式来额外配置

  #86 opened May 25, 2025

• 用户对话情感识别

  #83 opened May 23, 2025

3 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• 降噪功能应该精确到设备的配置中，如是否开启降噪，降噪强度设置等，并且后续可以通过自己通过识别周边噪音进行更新噪声阈值，需在设备编辑功能中修改

  #42 commented on May 25, 2025 • 0 new comments

• 历史对话页面目前只保存了Ai的对话语音（并且语音采用的edge生成的）考虑应该把生成的语音片段最终合并作为完整片段update到数据库中

  #43 commented on May 25, 2025 • 0 new comments

• 声纹识别

  #41 commented on May 25, 2025 • 0 new comments