Summary:
When you use the "Voice Parsing" feature, your recording is uploaded to our server, transcribed into text via a speech-to-text API, and then analyzed by a text AI model. The audio is processed in memory and is not stored on the server or any cloud storage after the request completes.
1. Information We Collect
1.1 Information You Provide
-
Audio data (recordings): When you trigger voice parsing, the app uploads your recorded
.m4a audio to our server for analysis.
-
User content (ledger entries): Amounts, titles, categories, and other bookkeeping data you enter or confirm are stored locally on your device (SwiftData).
1.2 Automatically Generated / Processed Information
-
Parsing results: The server cleans and structures the output returned by the third-party model and sends it back to the app.
-
Basic logs: To maintain service stability, the server may record necessary error logs (not for user profiling purposes).
2. How We Use Your Information
- Provide voice parsing: Convert audio into structured ledger entries (amount / title / category).
- Improve reliability: Diagnose upload failures, parsing errors, and similar issues.
3. How We Share Information (Third-Party Processing)
To deliver the voice parsing feature, we share your audio data with the following types of third-party service providers to the extent necessary:
-
Speech-to-text services: Your audio is sent to a speech recognition API to be transcribed into text.
-
AI text model services: The transcript is sent to a language model to extract structured ledger entries (example: Alibaba Cloud DashScope / Qwen).
Note: The names and locations of third-party services depend on your actual deployment and chosen providers. Please replace the "examples" above with your real service information before publishing.
4. Data Retention & Deletion
-
Local ledger data: Stored on your device. You can clear all data within the app.
-
Server-side audio: Audio is transmitted to our server solely for processing. It is passed directly to the speech-to-text API in memory and is not written to disk or stored in any cloud storage. Once the request completes, the audio data is discarded.
5. Your Choices & Controls
-
Skip voice parsing: You may choose to use only "Manual Entry," which requires no audio upload.
-
System permissions: You can revoke microphone access in iOS Settings, which will disable the recording feature.
-
Clear data: You can clear local ledger data in the app under Settings → Data Management.
6. Children's Privacy
This service is not directed at children. We do not knowingly collect personal information from children. If you believe we may have collected a child's information without appropriate consent, please contact us so we can address it.
7. Policy Updates
We may update this Privacy Policy from time to time. The updated version will be posted on this page with a revised effective date.
8. Contact Us
If you have any questions about this Privacy Policy or wish to exercise your rights, please contact us at:
一句话概览：
我们会在你使用"语音解析"功能时上传你的录音到服务器，服务端会调用语音转文字 API 将音频转为文字，再由文本 AI 模型解析为记账条目。音频在服务器内存中处理完毕后即被丢弃，不会存储到服务器本地或任何云存储中。
1. 我们收集哪些信息
1.1 你主动提供的信息
-
音频数据（录音）：当你使用 SmartLedger AI 并触发解析时，App 会上传你录制的
.m4a 音频到我们的服务器用于解析。
-
用户内容（记账条目）：你在 App 内输入或确认的金额、标题、分类等记账信息会保存在你的设备本地（SwiftData）。
1.2 自动生成/处理的信息
-
解析结果：服务器会对第三方模型返回的解析结果进行清洗/结构化，返回给 App 展示与保存。
-
基础日志：为保障服务稳定性，服务器可能记录必要的错误日志（不以建立用户画像为目的）。
2. 我们如何使用这些信息
- 提供语音解析服务：将音频解析为结构化的记账条目（金额/标题/分类）。
- 改进可靠性：用于排查上传失败、解析失败等问题。
3. 我们如何共享信息（第三方处理）
为了完成"语音解析"，我们会在必要范围内将你的音频数据共享给以下类型的第三方服务提供商进行处理：
-
语音转文字服务：你的音频会被发送至语音识别 API 转为文字。
-
AI 文本模型服务：转写后的文字会被发送至语言模型，提取结构化记账条目（示例：阿里云 DashScope / Qwen）。
说明：第三方服务的名称与所在地取决于你的实际部署与所选供应商；请在上架前将本页中的"示例"替换为你的真实信息。
4. 数据保存与删除
-
本地记账数据：保存在你的设备本地，你可在 App 内清除全部数据。
-
服务器端音频：音频仅用于处理，在服务器内存中直接传递给语音转文字 API，不会写入磁盘或上传至任何云存储。请求完成后，音频数据即被丢弃。
5. 你的选择与控制
-
不使用语音解析：你可以选择仅使用"手动记账"，无需上传音频。
-
系统权限：你可在 iOS 设置中关闭麦克风权限，但会影响录音功能。
-
清除数据：你可以在 App 的"设置 → 数据管理"中清除本地记账数据。
6. 儿童隐私
本服务不面向儿童，我们不会在明知的情况下收集儿童的个人信息。如你认为我们可能在未获得适当同意的情况下收集了儿童信息，请联系我们以便处理。
7. 隐私政策更新
我们可能会不定期更新本隐私政策。更新后的版本将发布在本页面，并更新"生效日期"。
8. 联系我们
如你对本隐私政策有任何疑问或需要行使相关权利，请通过以下方式联系我们：