VoiceSnip 把你在浏览器标签页里主动录制的声音,交给你自己配置的 AssemblyAI 账号去分离说话人,然后把片段和文字存到你自己电脑上。开发者 没有服务器,收不到你的录音、文字或密钥,也没有统计和广告。
| 数据 | 什么时候产生 | 用来做什么 | 存在哪里 |
|---|---|---|---|
| 标签页音频(16 kHz 单声道 WAV) | 只有你点「开始录制」,或按快捷键 / 用右键菜单主动发起时才录制 | 分离说话人、生成片段与配套文本 | 你浏览器的本地存储(OPFS),只保留最近 3 段,更早的自动删除 |
| AssemblyAI API 密钥 | 你自己在设置里粘贴并保存时 | 代表你调用你自己的 AssemblyAI 账号 | chrome.storage.local,只在本机,不写日志、不随导出文件外传 |
| 转写与分离结果 | 云端返回后 | 让你关掉面板再打开时不必重新上传、不重复付费 | chrome.storage.local,导出成功后即清除 |
| 设置与界面状态 | 你切换语言、同意本政策、选过说话人时 | 记住你的选择 | chrome.storage.local |
| 导出的音频与文本文件 | 你点导出时 | 交付产物 | 你自己电脑的下载目录 下载\voice-extract\,由你自行处置 |
只有一方:你自己配置的 AssemblyAI 账号(域名
api.assemblyai.com)。扩展把录音上传到你自己的账号做说话人分离与
转写,再取回结果。这条传输走 HTTPS。
AssemblyAI 是独立的第三方服务,它的数据处理由 AssemblyAI 自己的隐私政策 和你的账号设置决定。开发者无法访问、也无法代为删除你账号里的数据。如果你没有 配置密钥,扩展完全不联网,只在本机按静音切分音频。
所有网络请求均使用 HTTPS;密钥只存放在本机
chrome.storage.local,不会写入日志,也不会出现在导出的音频或文本文件里。
扩展不加载、也不执行任何远程下发的代码。
从 Google API 获得的信息的使用将遵守 Chrome 应用商店用户数据政策,包括其中的有限使用(Limited Use)要求。
本扩展面向能够自行理解并同意本政策的成年用户,不面向儿童,也不会有意收集儿童的 任何数据。
数据做法有实质变化时,我们会更新本页并修改顶部的生效日期;如果变化影响到你, 扩展会在你下次打开时重新征求你的同意。
任何与本政策有关的问题,写信到 bright103huang@gmail.com。
VoiceSnip records audio from a browser tab that you explicitly start recording, hands that audio to the AssemblyAI account you configured so it can separate the speakers, and then saves the clips and transcripts on your own computer. The developer runs no servers and never receives your audio, transcripts or API key. There is no analytics and no advertising.
| Data | When it is produced | Why | Where it lives |
|---|---|---|---|
| Tab audio (16 kHz mono WAV) | Only after you press Start recording, press the shortcut, or use the context menu entry yourself | To separate speakers and produce clips plus transcripts | Your browser's local storage (OPFS); only the last 3 takes are kept, older ones are deleted automatically |
| AssemblyAI API key | When you paste and save it in the settings panel | To call your own AssemblyAI account on your behalf | chrome.storage.local, on this machine only. Never logged,
never included in exported files |
| Transcription and separation results | After the cloud returns them | So reopening the panel does not re-upload or charge you twice | chrome.storage.local, cleared as soon as you export |
| Settings and UI state | When you switch language, accept this policy, or pick a speaker | To remember your choices | chrome.storage.local |
| Exported audio and text files | When you press export | To deliver the result | Your own Downloads folder, voice-extract\, under your control |
Exactly one party: the AssemblyAI account you configured yourself
(api.assemblyai.com). The extension uploads the recording to your own
account for speaker separation and transcription, then retrieves the result. That
transfer uses HTTPS.
AssemblyAI is an independent third-party service, and its handling of your data is governed by AssemblyAI's own privacy policy and your account settings. The developer cannot access or delete data inside your account. If you have not configured a key, the extension makes no network requests at all and splits audio locally by silence.
All network requests use HTTPS. The key is kept only in local
chrome.storage.local, is never written to logs, and never appears in
exported audio or text files. The extension does not load or execute any remotely
hosted code.
The use of information received from Google APIs will adhere to the Chrome Web Store User Data Policy, including the Limited Use requirements.
This extension is intended for adult users who can read and agree to this policy. It is not directed at children and does not knowingly collect any data from them.
If our data practices change materially we will update this page and the effective date at the top. If a change affects you, the extension will ask for your consent again the next time you open it.
For any question about this policy, write to bright103huang@gmail.com.