| 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156 |
- # 应用配置
- app:
- name: shudao-chat-py
- host: 0.0.0.0
- port: 22001
- debug: true
- # MySQL数据库配置
- database:
- user: root
- password: "88888888"
- host: 172.16.29.101
- port: 21000
- database: shudao
- pool_size: 100
- max_overflow: 10
- pool_recycle: 3600
- # DeepSeek配置
- deepseek:
- api_key: sk-28625cb3738844e190cee62b2bcb25bf
- api_url: https://api.deepseek.com
- # 本地大模型配置
- # qwen3:
- # api_url: http://172.16.35.50:8020
- # model: /opt/Qwen3.5-122B-A10B
- qwen3:
- api_url: https://aigc-api.aitoolcore.com
- model: qwen3.6-plus
- token: sk-aigc-06515b42c1522bdb7f2473672ef6363c553e5a4a
- # 意图识别模型配置
- # intent:
- # api_url: http://172.16.35.50:8020
- # model: /opt/Qwen3.5-122B-A10B
- intent:
- api_url: https://aigc-api.aitoolcore.com
- model: qwen3.6-plus
- token: sk-aigc-06515b42c1522bdb7f2473672ef6363c553e5a4a
-
- # YOLO API配置
- yolo:
- base_url: http://172.16.35.50:18080
- # Dify Workflow配置
- dify:
- workflow_url: http://172.16.35.50:8000/v1/workflows/run
- workflow_id: 4wfh1PPDderMtCeb
- auth_token: app-55CyO4lmDv1VeXK4QmFpt4ng
- # AIChat 服务配置
- aichat:
- api_url: http://127.0.0.1:28002/api/v1
- timeout: 600
- # 基础URL配置
- base_url: https://aqai.shudaodsj.com:22000
- # Token验证API配置
- auth:
- api_url: http://127.0.0.1:28004/api/auth/verify
- # OSS配置
- oss:
- access_key_id: fnyfi2f368pbic74d8ll
- access_key_secret: jgqwk7sirqlz2602x2k7yx2eor0vii19wah6ywlv
- bucket: gdsc-ai-aqzs
- endpoint: http://172.16.17.52:8060
- parse_encrypt_key: jgqwk7sirqlz2602
- # 思考过程二次总结（方案三）：不展示原始<think>长文，仅输出中文要点
- thinking_summary:
- enabled: true
- max_points: 5
- max_input_chars: 1500
- max_output_chars: 600
- temperature: 0.2
- # 语音服务配置（方案1：腾讯云直连后端，挂载在 shudao-chat-py）
- speech:
- enabled: true
- provider: tencent_cloud
- integration_mode: backend_direct
- backend_service: shudao-chat-py
- request_timeout_seconds: 60
- # 需要申请/开通的第三方能力
- procurement:
- required_cloud_products:
- - asr_sentence_recognition
- - asr_flash_file_recognition
- - tts_text_to_voice
- - tts_long_text
- required_credentials:
- - app_id
- - secret_id
- - secret_key
- optional_cloud_products:
- - cos
- notes:
- - sentence_recognition_for_audio_within_60s_and_3mb
- - flash_file_recognition_for_longer_audio_fallback
- - text_to_voice_for_default_tts
- - long_text_tts_for_content_over_150_chinese_chars
- # 腾讯云账号与接口入口
- tencent:
- app_id: "1258082136"
- secret_id: "AKIDtgFYdS2OVO1PZT8XfZEd7LK3XmlUyDDy"
- secret_key: "XoqdASfdMuEKAk5B2e1DArJXVDfsqY2G"
- region: "ap-chengdu"
- asr_endpoint: "asr.tencentcloudapi.com"
- tts_endpoint: "tts.tencentcloudapi.com"
- # 语音转文字：第一版优先一句话识别，超出范围时切录音文件识别极速版
- transcribe:
- enabled: true
- default_api: "SentenceRecognition"
- fallback_api: "CreateRecTask"
- engine_model_type: "16k_zh"
- source_type: 1
- voice_format: "wav"
- input_sample_rate: 16000
- max_audio_seconds: 60
- max_audio_size_mb: 3
- word_info: 0
- filter_dirty: 0
- filter_modal: 0
- filter_punc: 0
- convert_num_mode: 1
- hotword_list: ""
- hotword_id: ""
- customization_id: ""
- replace_text_id: ""
- # 语音播报：第一版优先基础语音合成，长文本时切长文本异步合成
- synthesize:
- enabled: true
- default_api: "TextToVoice"
- fallback_api: "CreateTtsTask"
- primary_language: 1
- voice_type: 1001
- sample_rate: 16000
- codec: "mp3"
- speed: 0.0
- volume: 0.0
- enable_subtitle: false
- segment_rate: 0
- emotion_category: ""
- emotion_intensity: 100
- basic_text_limit_chars: 150
- long_text_trigger_chars: 120
- long_text_max_chars: 100000
- long_text_codec: "mp3"
- callback_url: ""
|
