Back to Details

sales-wave

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Wave Platform Help

Wave平台帮助

Step 1 — Gather context

步骤1 — 收集上下文

If 
references/learnings.md
exists, read it first for accumulated platform knowledge.
  1. What do you need help with?
    • A) Recording setup — getting Wave to record calls, meetings, or lectures reliably
    • B) API integration — pulling transcripts/summaries into another system
    • C) Webhooks — real-time notifications when sessions complete
    • D) Export/sharing — getting notes into Notion, Google Docs, OneNote, etc.
    • E) Pricing/plan selection — which Wave tier is right for my usage
    • F) Troubleshooting — something isn't working as expected
  2. How are you recording?
    • A) Mobile app (iPhone/Android) — in-person meetings or phone calls
    • B) Desktop app (Mac/Windows) — system audio capture
    • C) Meeting bot — Zoom/Meet/Teams calendar integration
    • D) Chrome extension — browser-based meetings
    • E) Apple Watch
    • F) Multiple methods
Skip-ahead rule: if the user's prompt already contains enough context, skip to Step 2.
如果
references/learnings.md
文件存在，请先阅读以获取累积的平台知识。
  1. 您需要哪方面的帮助？
    • A) 录音设置——确保Wave可靠录制通话、会议或讲座
    • B) API集成——将转录内容/摘要同步至其他系统
    • C) Webhooks配置——会话完成时获取实时通知
    • D) 导出/分享——将笔记导入Notion、Google Docs、OneNote等平台
    • E) 定价/方案选择——哪款Wave套餐适合您的使用需求
    • F) 故障排查——功能未按预期工作
  2. 您使用哪种录音方式？
    • A) 移动端应用（iPhone/Android）——用于线下会议或电话通话
    • B) 桌面端应用（Mac/Windows）——捕获系统音频
    • C) 会议机器人——Zoom/Meet/Teams日历集成
    • D) Chrome扩展程序——基于浏览器的会议
    • E) Apple Watch
    • F) 多种方式
跳过规则：如果用户的提问已包含足够上下文，直接进入步骤2。

Step 2 — Route or answer directly

步骤2 — 转至对应流程或直接解答

Problem domainRoute to
Choosing between Wave and other note-takers
/sales-note-taker {user's question}
Reviewing a specific call for coaching
/sales-call-review {user's question}
Building a coaching program from call data
/sales-coaching {user's question}
General CRM/tool integration (Zapier, iPaaS)
/sales-integration {user's question}
Otherwise, answer directly below.
问题领域转至路径
在Wave与其他笔记工具间做选择
/sales-note-taker {用户问题}
为辅导目的复盘特定通话
/sales-call-review {用户问题}
基于通话数据构建辅导体系
/sales-coaching {用户问题}
通用CRM/工具集成（Zapier、iPaaS）
/sales-integration {用户问题}
若不属于上述情况，请直接在下方解答。

Step 3 — Wave platform reference

步骤3 — Wave平台参考资料

Read 
references/platform-guide.md
 for the full platform reference — recording methods, pricing, integrations, Voice ID, export options, and common workflows.
Read 
references/wave-api-reference.md
 for API endpoints, authentication, webhooks, rate limits, and the OpenAPI spec.
Answer the user's question using only the relevant section. Don't dump the full reference.
**阅读
references/platform-guide.md
**获取完整平台参考——包含录音方式、定价、集成功能、Voice ID、导出选项及常见工作流程。
**阅读
references/wave-api-reference.md
**获取API端点、认证方式、webhooks、速率限制及OpenAPI规范。
仅使用相关章节内容解答用户问题，请勿直接粘贴完整参考资料。

Step 4 — Actionable guidance

步骤4 — 可落地指导

You no longer need the platform guide — focus on the user's specific situation.
Recording reliability checklist:
  1. Check which recording method matches the use case (mobile mic for in-person, desktop for system audio, bot for scheduled virtual meetings)
  2. For phone calls on iPhone, verify Phone Bridge is enabled and working
  3. For meeting bots, confirm calendar integration is connected and auto-join is enabled
  4. Test with a short recording before relying on it for important calls
API integration pattern:
  1. Create token at 
    app.wave.co/settings/integrations/api
    with needed scopes
  2. Register a webhook for 
    session.completed
    events to get real-time notifications
  3. On webhook fire, call 
    GET /v1/sessions/{id}/transcript
    for full transcript with speaker segments
  4. Store raw JSON, then normalize into your common schema
If you discover a gotcha, workaround, or tip not covered in 
references/learnings.md
, append it there.
无需再参考平台指南——专注于用户的具体场景。
录音可靠性检查清单：
  1. 确认录音方式与使用场景匹配（线下会议用移动端麦克风，系统音频用桌面端，预定虚拟会议用机器人）
  2. 对于iPhone通话，验证Phone Bridge已启用且正常工作
  3. 对于会议机器人，确认日历集成已连接且自动加入功能已开启
  4. 在重要通话前先测试短时长录音
API集成模式：
  1. 在
    app.wave.co/settings/integrations/api
    创建具有所需权限范围的令牌
  2. 注册
    session.completed
    事件的webhook以获取实时通知
  3. webhook触发时，调用
    GET /v1/sessions/{id}/transcript
    获取带说话人分段的完整转录内容
  4. 存储原始JSON数据，然后标准化为您的通用 schema
如果发现
references/learnings.md
未涵盖的注意事项、解决方案或技巧，请将其追加至该文件。

Gotchas

常见陷阱

Best-effort from research — review these, especially items about plan-gated features and integration gotchas that may be outdated.
  • Incoming calls kill active recordings on mobile. If a phone call comes in while Wave is recording, the recording stops and prior audio may be lost. No resume/append capability. Use Do Not Disturb mode during important recordings.
  • Free tier is only 30 minutes/month. Enough for a test, not for regular use. Pro ($11.67/mo annual) is effectively the entry point.
  • API tokens expire after 1 year. Set a calendar reminder to rotate — failed auth will silently stop your pipeline.
  • Media URLs expire after 1 hour. If you need to store audio/video, download immediately after fetching the signed URL. Don't cache the URL itself.
  • Rate limits are tight for bulk operations. 60 req/min and 1,000/day per token. For backfills, use 
    POST /v1/sessions/bulk
    (up to 50 sessions) instead of individual fetches.
  • Voice ID requires training. Speaker labeling improves over time as Wave learns voices — early recordings may have unlabeled or mislabeled speakers.
  • "Stuck on saving" is a known issue. Some users report the app hanging on transcription. Force-close and reopen — the recording is usually saved but processing may be delayed.
  • Export formatting degrades in OneNote. Copy-paste from Wave to OneNote loses formatting. Use the native OneNote integration instead of clipboard.
基于研究的最佳实践——请重点关注这些内容，尤其是关于套餐专属功能和可能过时的集成陷阱项。
  • 移动端来电会中断当前录音：如果Wave录音时有来电，录音会停止，之前的音频可能丢失。不支持恢复/追加功能。重要录音期间请开启勿扰模式。
  • 免费套餐每月仅支持30分钟：足够用于测试，但无法满足常规使用需求。专业版（年付每月$11.67）是实际的入门选择。
  • API令牌1年后过期：设置日历提醒进行轮换——认证失败会导致您的流程静默停止。
  • 媒体URL 1小时后过期：如果需要存储音频/视频，请在获取签名URL后立即下载。请勿缓存URL本身。
  • 批量操作的速率限制严格：每个令牌每分钟60次请求，每天1000次请求。对于批量回填，使用
    POST /v1/sessions/bulk
    （最多支持50个会话）而非单个请求获取。
  • Voice ID需要训练：随着Wave学习说话人声音，标记准确率会逐步提升——早期录音可能出现未标记或标记错误的情况。
  • “保存中”是已知问题：部分用户反馈应用在转录时卡顿。强制关闭并重新打开应用——录音通常已保存，但处理可能延迟。
  • 导出至OneNote时格式会丢失：从Wave复制粘贴至OneNote会丢失格式，请使用原生OneNote集成而非剪贴板。

Related skills

相关技能

  • /sales-note-taker
    — Comparing AI note-takers (Fathom, Fireflies, Avoma, Gong, Otter, and 10+ others) or building a transcript pipeline across vendors
  • /sales-call-review
    — Review a specific call for coaching
  • /sales-coaching
    — Build coaching programs that consume call data
  • /sales-integration
    — General CRM/tool integration patterns (Zapier, webhooks, iPaaS)
  • /sales-do
    — Not sure which skill to use? The router matches any sales objective to the right skill. Install: 
    npx skills add sales-skills/sales --skill sales-do
  • /sales-note-taker
    ——对比AI笔记工具（Fathom、Fireflies、Avoma、Gong、Otter及10余款其他工具）或构建跨供应商的转录流程
  • /sales-call-review
    ——为辅导目的复盘特定通话
  • /sales-coaching
    ——构建基于通话数据的辅导体系
  • /sales-integration
    ——通用CRM/工具集成模式（Zapier、webhooks、iPaaS）
  • /sales-do
    ——不确定使用哪项技能？该路由会将任何销售目标匹配至合适技能。安装方式：
    npx skills add sales-skills/sales --skill sales-do

Examples

示例

Example 1: Recording setup for a sales team

示例1：销售团队录音设置

User says: "I want my 4-person sales team to use Wave for all our Zoom calls. How do I set it up so recordings auto-start?" Skill does: Walks through calendar integration for meeting bot auto-join, recommends Teams plan ($7.50/user/mo), explains shared workspace for team visibility, notes Voice ID training for speaker attribution.
用户提问：“我希望我的4人销售团队使用Wave录制所有Zoom通话。如何设置让录音自动启动？” 技能操作：引导完成会议机器人自动加入的日历集成配置，推荐Teams套餐（每位用户每月$7.50），讲解团队可见的共享工作区，说明说话人归因的Voice ID训练方法。

Example 2: Piping transcripts to HubSpot

示例2：将转录内容同步至HubSpot

User says: "How do I automatically get Wave transcripts into HubSpot after every call?" Skill does: Outlines webhook setup for 
session.completed
, API call to fetch transcript, then either Zapier (simple) or custom middleware (flexible) to push to HubSpot call activity. Notes rate limits and token scoping.
用户提问：“如何在每次通话后自动将Wave转录内容同步至HubSpot？” 技能操作：概述
session.completed
的webhook设置流程、获取转录内容的API调用方法，然后介绍使用Zapier（简易方案）或自定义中间件（灵活方案）推送至HubSpot通话活动。提醒速率限制和令牌权限范围。

Example 3: Recording stops mid-call

示例3：录音中途停止

User says: "Wave recorded only the first 5 minutes of my 45-minute call. What happened?" Skill does: Checks if an incoming call interrupted recording (most common cause), whether battery optimization killed the app in background, or if storage was full. Recommends Do Not Disturb mode and testing with a short call first.
用户提问：“Wave只录制了我45分钟通话的前5分钟。是什么原因？” 技能操作：检查是否有来电中断录音（最常见原因）、电池优化是否在后台关闭了应用、或存储是否已满。建议开启勿扰模式，并先通过短时长通话测试可靠性。

Troubleshooting

故障排查

Recording stops or captures only a few minutes

录音停止或仅录制几分钟

Symptom: Wave records the beginning of a call but stops partway through Cause: Most commonly, an incoming phone call interrupts the recording. Also: battery optimization killing the app, storage full, or app crash during long recordings Solution: Enable Do Not Disturb before important recordings. On Android, exempt Wave from battery optimization. Check available storage. For calls >1 hour, test reliability first with a shorter recording.
症状：Wave录制了通话开头部分，但中途停止 原因：最常见的是来电中断录音。其他原因包括：电池优化关闭应用、存储已满、或长时间录音时应用崩溃 解决方案：重要录音前开启勿扰模式。在Android设备上，将Wave排除在电池优化之外。检查可用存储空间。对于超过1小时的通话，先通过短时长测试确认可靠性。

App stuck on "saving recording" or transcription hangs

应用卡在“保存录音”或转录卡顿

Symptom: After recording, the app shows "saving" or processing spinner indefinitely Cause: Large recordings can take time to process; network interruptions during upload; known intermittent bug Solution: Wait 5-10 minutes. If still stuck, force-close and reopen — the recording is usually saved server-side. Check internet connection. If the recording is lost, contact support@wave.co.
症状：录音后，应用显示“保存中”或处理加载图标持续显示 原因：大文件录音需要较长处理时间；上传时网络中断；已知间歇性bug 解决方案：等待5-10分钟。如果仍未恢复，强制关闭并重新打开应用——录音通常已保存在服务器端。检查网络连接。如果录音丢失，请联系support@wave.co。

Webhook fires but transcript has no speaker labels

Webhook触发但转录内容无说话人标记

Symptom: 
session.completed
webhook payload arrives, transcript text is present but all segments show "Unknown Speaker" Cause: Voice ID hasn't been trained for these speakers yet, or the recording quality was too low for speaker diarization Solution: Voice ID improves with use — label speakers manually in the first few recordings to train the model. For virtual meetings, meeting bot recordings have better speaker separation than mobile mic recordings.
症状：
session.completed
webhook payload已接收，转录文本存在但所有分段显示“Unknown Speaker” 原因：Voice ID尚未针对这些说话人完成训练，或录音质量过低导致说话人分离失败 解决方案：Voice ID会随使用逐步优化——在最初几次录音中手动标记说话人以训练模型。对于虚拟会议，会议机器人录制的内容比移动端麦克风录制的内容说话人分离效果更好。