# 火山 Websocket 双向流式对话 ## 例程简介 本例程实现了扣子智能语音对话 Websocket OpenAPI， 主要是通过按键的方式实现与智能体的对讲。 ## 示例创建 ### IDF 默认分支 本例程支持 IDF release/v5.4 及以后的分支。 ### 预备知识 首先需要在[Coze文档中](https://bytedance.larkoffice.com/docx/Da6qd87pQodvNrxdFYrcnzMxnsh)申请 `Access token` 和 `BOT ID`账号 更多的 Websocket 文档可以参考 [双向流式对话事件](https://www.coze.cn/open/docs/developer_guides/streaming_chat_event) ### 配置 1. 将获取到的 `Access token` 和 `BOT ID` 信息填入 `Menuconfig->Example Configuration` 中。 2. 將 wifi 信息填入 `Menuconfig->>Example Configuration` 中。 ### 编译和下载 编译本例程前需要先确保已配置 ESP-IDF 的环境，如果已配置可跳到下一项配置，如果未配置需要先在 ESP-IDF 根目录运行下面脚本设置编译环境，有关配置和使用 ESP-IDF 完整步骤，请参阅 [《ESP-IDF 编程指南》](https://docs.espressif.com/projects/esp-idf/zh_CN/latest/esp32s3/index.html) ``` ./install.sh . ./export.sh ``` - 选择编译芯片，以 esp32s3 为例： ``` idf.py set-target esp32s3 ``` - 编译例子程序 ``` idf.py build ``` - 烧录程序并运行 monitor 工具来查看串口输出 (替换 PORT 为端口名称)： ``` idf.py -p PORT flash monitor ``` - 退出调试界面使用 ``Ctrl-]`` ## 如何使用例程 ### 功能和用法 - 例程开始运行后， 当出现以下log就说明了与服务端建立了连接， 就可以对话了: ```c I (931) main: Initialize board peripherals W (932) i2c_bus_v2: I2C master handle is NULL, will create new one I (941) DRV8311: ES8311 in Slave mode I (958) ES7210: ES7210 in Slave mode I (967) ES7210: Enable ES7210_INPUT_MIC1 I (970) ES7210: Enable ES7210_INPUT_MIC2 I (972) ES7210: Enable ES7210_INPUT_MIC3 W (976) ES7210: Enable TDM mode. ES7210_SDP_INTERFACE2_REG12: 2 I (980) ES7210: config fmt 60 I (982) AUDIO_HAL: Codec mode is 3, Ctrl:1 I (990) pp: pp rom version: e7ae62f I (990) net80211: net80211 rom version: e7ae62f I (990) AUDIO_THREAD: The esp_periph task allocate stack on internal memory I (991) wifi:wifi driver task: 3fcee994, prio:23, stack:6656, core=0 I (1001) wifi:wifi firmware version: 21fc8af6de I (1003) wifi:wifi certification version: v7.0 I (1007) wifi:config NVS flash: enabled I (1011) wifi:config nano formatting: disabled I (1015) wifi:Init data frame dynamic rx buffer num: 32 I (1020) wifi:Init static rx mgmt buffer num: 5 I (1024) wifi:Init management short buffer num: 32 I (1029) wifi:Init dynamic tx buffer num: 32 I (1033) wifi:Init static tx FG buffer num: 2 I (1037) wifi:Init static rx buffer size: 1600 I (1041) wifi:Init static rx buffer num: 16 I (1045) wifi:Init dynamic rx buffer num: 32 I (1049) wifi_init: rx ba win: 16 I (1052) wifi_init: accept mbox: 32 I (1055) wifi_init: tcpip mbox: 32 I (1058) wifi_init: udp mbox: 6 I (1061) wifi_init: tcp mbox: 32 I (1064) wifi_init: tcp tx win: 65535 I (1067) wifi_init: tcp rx win: 32768 I (1071) wifi_init: tcp mss: 1440 I (1074) wifi_init: WiFi IRAM OP enabled I (1078) wifi_init: WiFi RX IRAM OP enabled W (1082) wifi:Password length matches WPA2 standards, authmode threshold changes from OPEN to WPA2 I (1090) wifi:Set ps type: 1, coexist: 0 I (1094) phy_init: phy_version 701,f4f1da3a,Mar 3 2025,15:50:10 I (1155) wifi:mode : sta (74:4d:bd:9d:b6:30) I (1155) wifi:enable tsf W (1155) PERIPH_WIFI: WiFi Event cb, Unhandle event_base:WIFI_EVENT, event_id:43 I (2446) wifi:new:<11,0>, old:<1,0>, ap:<255,255>, sta:<11,0>, prof:1, snd_ch_cfg:0x0 I (2446) wifi:state: init -> auth (0xb0) W (2447) PERIPH_WIFI: WiFi Event cb, Unhandle event_base:WIFI_EVENT, event_id:43 I (2462) wifi:state: auth -> assoc (0x0) I (2468) wifi:state: assoc -> run (0x10) I (2485) wifi:connected with xtworks, aid = 88, channel 11, BW20, bssid = ec:56:23:e9:7e:f0 I (2486) wifi:security: WPA2-PSK, phy: bgn, rssi: -40 I (2488) wifi:pm start, type: 1 I (2490) wifi:dp: 1, bi: 102400, li: 3, scale listen interval from 307200 us to 307200 us I (2498) wifi:set rx beacon pti, rx_bcn_pti: 0, bcn_timeout: 25000, mt_pti: 0, mt_time: 10000 W (2508) PERIPH_WIFI: WiFi Event cb, Unhandle event_base:WIFI_EVENT, event_id:4 I (2532) wifi:AP's beacon interval = 102400 us, DTIM period = 1 I (3571) esp_netif_handlers: sta ip: 192.168.3.7, mask: 255.255.255.0, gw: 192.168.3.1 I (3571) PERIPH_WIFI: Got ip:192.168.3.7 I (3571) coze_chat: wss_url: ws://ws.coze.cn/v1/chat?bot_id=74800678******** I (3579) AUDIO_THREAD: The coze_data_pull_task task allocate stack on external memory I (3586) AUDIO_THREAD: The coze_data_push_task task allocate stack on external memory I (3594) coze_chat: WEBSOCKET_EVENT_BEGIN I (3598) websocket_client: Started I (3600) coze_chat: Wait for websocket connected I (4104) coze_chat: Wait for websocket connected I (4192) coze_chat: WEBSOCKET_EVENT_CONNECTED I (5215) wifi:idx:0 (ifx:0, ec:56:23:e9:7e:f0), tid:6, ssn:2665, winSize:64 I (5265) coze_chat: Request conversation_id : 7491159********* I (5266) coze_chat: WS connected I (5266) coze_chat: WS updata chat I (5268) coze_chat: Update chat: { "id": "15deeac7-bc4c-731a-c81e-2585128c3557", "event_type": "chat.update", "data": { "chat_config": { "auto_save_history": true, "conversation_id": "7491159*********", "user_id": "userid_123", "meta_data": { }, "custom_variables": { }, "extra_params": { }, "parameters": { "custom_var_1": "测试" } }, "input_audio": { "format": "pcm", "codec": "pcm", "sample_rate": 16000, "channel": 1, "bit_depth": 16 }, "turn_detection": { "type": "server_vad", "prefix_padding_ms": 600, "silence_duration_ms": 500 }, "output_audio": { "codec": "opus", "opus_config": { "bitrate": 16000, "frame_size_ms": 60, "limit_config": { "period": 1, "max_frame_num": 18 } }, "speech_rate": 20, "voice_id": "7426720361733144585" }, "event_subscriptions": ["conversation.audio.delta", "conversation.chat.completed", "input_audio_buffer.speech_started", "input_audio_buffer.speech_stopped", "chat.created", "error"] } } I (5365) MODEL_LOADER: The storage free size is 23104 KB I (5367) MODEL_LOADER: The partition size is 5168 KB I (5372) MODEL_LOADER: Successfully load srmodels I (5376) ALGORITHM_STREAM: Load: wn9_hilexin I (5381) AUDIO_PIPELINE: link el->rb, el:0x3c17e144, tag:algo_stream, rb:0x3c17e3d0 I (5389) AUDIO_PIPELINE: link el->rb, el:0x3c185108, tag:raw_stream, rb:0x3c185538 I (5395) AUDIO_PIPELINE: link el->rb, el:0x3c185250, tag:raw_opus, rb:0x3c187580 I (5402) AUDIO_THREAD: The algo_stream task allocate stack on external memory I (5409) AUDIO_ELEMENT: [algo_stream-0x3c17e144] Element task created I (5415) AUDIO_ELEMENT: [raw_read-0x3c17e274] Element task created I (5421) AUDIO_PIPELINE: Func:audio_pipeline_run, Line:359, MEM Total:8447912 Bytes, Inter:201343 Bytes, Dram:201343 Bytes, Dram largest free:102400Bytes I (5435) AUDIO_ELEMENT: [algo_stream] AEL_MSG_CMD_RESUME,state:1 I (5440) AUDIO_PIPELINE: Pipeline started I (5440) AFE_CONFIG: Set WakeNet Model: wn9_hilexin I (5444) AUDIO_ELEMENT: [raw_stream-0x3c185108] Element task created W (5450) AFE_CONFIG: For single microphone channel, SE is deactivated. I (5456) AUDIO_THREAD: The raw_opus task allocate stack on external memory I (5485) AFE: AFE Version: (1MIC_V250121) I (5485) AFE: Input PCM Config: total 2 channels(1 microphone, 1 playback), sample rate:16000 I (5488) AFE: AFE Pipeline: [input] -> |AEC(VOIP_LOW_COST)| -> |NS(WebRTC)| -> [output] I (5497) AUDIO_THREAD: The algo_fetch task allocate stack on external memory I (5522) AUDIO_ELEMENT: [raw_opus-0x3c185250] Element task created I (5522) AUDIO_THREAD: The filter task allocate stack on external memory I (5523) AUDIO_ELEMENT: [filter-0x3c1853c8] Element task created I (5529) AUDIO_PIPELINE: Func:audio_pipeline_run, Line:359, MEM Total:8219944 Bytes, Inter:172823 Bytes, Dram:172823 Bytes, Dram largest free:98304Bytes I (5542) AUDIO_ELEMENT: [raw_opus] AEL_MSG_CMD_RESUME,state:1 I (5557) AUDIO_ELEMENT: [filter] AEL_MSG_CMD_RESUME,state:1 I (5557) AUDIO_PIPELINE: Pipeline started I (5558) AUDIO_THREAD: The audio_data_read_task task allocate stack on external memory I (5566) main: Func:app_main, Line:161, MEM Total:8187216 Bytes, Inter:171507 Bytes, Dram:171507 Bytes, Dram largest free:98304Bytes I (5582) main_task: Returned from app_main() ```