Listen to your script in headphones. Echo it to the camera. Only your voice ends up in the video.
戴上耳机听稿子，对着镜头复述一遍。视频里只留下你自己的声音。
Last updated: May 7, 2026
VoiceCue is designed with a simple principle: what you record is yours alone. The app does not collect, transmit, or share any personal information.
Nothing. VoiceCue does not collect, store, or transmit any personal data, usage data, analytics, or telemetry.
The app stores the following on your device only, never leaving it:
The on-device items are saved inside the app's sandbox and removed when you uninstall VoiceCue. The iCloud-mirrored install date is removed when you sign out of iCloud or reset your iCloud data.
VoiceCue asks for the following iOS permissions, purely for on-device work:
VoiceCue works fully offline. Text-to-speech runs on your iPhone via Apple's AVSpeechSynthesizer. Smart Edit's filler-word detection also runs entirely on-device, using bundled Whisper-tiny (CoreML) and SenseVoice (ONNX) speech-recognition models that ship inside the app — your audio is transcribed locally and never uploaded to any server for processing. No audio, video, or text leaves your device.
None. VoiceCue does not use any third-party analytics, advertising, crash reporting, or tracking services.
VoiceCue offers a one-time "Pro" purchase via Apple's StoreKit. Apple processes the payment; VoiceCue only receives a cryptographic receipt. We never see your Apple ID, payment details, or billing address.
Most future Pro feature updates will be included free for existing buyers. However, if a future feature requires paid third-party services (e.g., premium beauty filter SDK licensing, metered cloud speech recognition APIs), it may be offered as a separate higher-tier purchase or subscription rather than included in the original Pro tier.
VoiceCue is safe for all ages. Because no data is collected, no special provisions are needed for users under 13.
If this policy ever changes, the updated date at the top of this page will reflect it.
Questions? Reach out via any of these channels — I respond personally.
最后更新：2026 年 5 月 7 日
VoiceCue 的原则只有一条：你录下的东西只属于你。应用不收集、不上传、不分享任何个人信息。
一条都不收。VoiceCue 不会收集、存储或上传任何个人数据、使用数据、分析数据或遥测数据。
以下内容只保存在你自己的设备上，永远不会离开：
本地的内容保存在应用沙盒里，卸载 VoiceCue 时会一并清除。iCloud 里同步的那份起始日期，在你登出 iCloud 或重置 iCloud 数据时会被清除。
VoiceCue 会申请下面这几项 iOS 权限，全部用于设备本地功能：
VoiceCue 完全离线运行。文本转语音由 iPhone 本地的 Apple AVSpeechSynthesizer 完成。智能剪辑的口水词识别也是本地运行的，使用 app 内置的 Whisper-tiny（CoreML）和 SenseVoice（ONNX）语音识别模型直接在设备上转写你的音频，不会上传到任何服务器做处理。任何音频、视频、文字都不会离开你的设备。
没有。VoiceCue 不使用任何第三方分析、广告、崩溃统计或追踪服务。
VoiceCue 通过 Apple StoreKit 提供一次性买断的「完整版」购买。付款由 Apple 处理，VoiceCue 只接收加密确认凭证，我们不会获取你的 Apple ID、支付详情或账单地址。
大部分未来 Pro 功能更新将免费提供给已购买用户。但如果未来某些功能需要依赖付费的第三方服务（例如：高级美颜 SDK 的年度授权费、云端语音识别 API 的按量计费），我们可能会将这些特定功能作为单独的高级订阅或新一档买断提供，不强制涵盖在原 Pro 内。
VoiceCue 适合所有年龄段。因为应用不收集任何数据，所以不需要为 13 岁以下用户做特别处理。
如果本政策有变更，更新日期会显示在本页顶部。
有问题？通过下面任一方式联系我，我本人会回复。
No. The reference plays only in your headphones. The microphone captures your voice alone. The one thing to remember: always wear headphones.
Wired earphones give zero latency and the best feel. AirPods work well too. Other Bluetooth headphones may introduce slight playback lag — it can nudge your rhythm but it never contaminates the recording.
Technically yes, but the reference audio from the speaker will bleed into the microphone and muddy your recording. Headphones are the whole point — don't skip them.
Three ways, depending on what's easiest for you:
mp3, m4a, or wav from your phone.When you tap Save to Photos on the preview screen, the video goes to the iOS Photos app. You can also reopen recent takes from the app's library without saving them to Photos first.
Smart Edit watches your recording for two things: silent pauses (you stopping to think) and filler words (um, uh, 嗯, 啊, etc.). It marks them on the timeline so you can drop them with a tap before exporting. Everything happens on your phone — your audio is transcribed locally with bundled speech models and never leaves the device.
Recording, importing reference audio, saving to Photos, and 0.7×–1.3× playback speed are free forever. The 7-day free trial unlocks two extra features that take real time and money to build and ship — AI text-to-speech and Smart Edit. After the trial, those two stay locked behind a one-time purchase (no subscription, yours forever). Everything else keeps working untouched.
Yes. Speed adjusts from 0.7× to 1.3× so you can slow a fast reference down or push it a little. Aspect ratio offers 9:16 (portrait, for Reels / TikTok / Shorts), 3:4, and 1:1.
No. Every feature — including text-to-speech and Smart Edit's speech recognition — runs on your iPhone.
Reach out on any of these — I respond personally.
不会。参考音只在耳机里播放。麦克风只录你自己的声音。唯一要记住的事：戴耳机。
有线耳机零延迟，体验最好。AirPods 也可以。其他蓝牙耳机可能有轻微播放延迟，会影响你跟读的节奏，但不会影响录音质量。
技术上能，但外放的参考音会被麦克风录进去，成片会很混乱。耳机是这个方法的核心，别省。
三种方式，选你最方便的那一种：
mp3、m4a、wav 文件都能用。在预览页点保存到相册，视频会写入 iOS 相册。没保存也没关系，可以在应用的我的录音里重新打开最近录的视频。
智能剪辑会在你录完之后扫一遍片子里的两种东西：静音停顿（你思考时的间隙）和口水词（嗯、啊、然后、um、uh 这些）。它把它们标在时间轴上，你可以一键剔掉再导出。整个过程都在手机本地，音频用 app 自带的语音识别模型转写，永远不会离开你的设备。
录视频、导入参考音、保存到相册、0.7×–1.3× 倍速调节永久免费。7 天免费试用解锁的是两项做起来比较费工的功能——AI 文本转语音和智能剪辑。试用结束后这两项需要一次性买断（不是订阅，付一次永久解锁），其他功能照常使用，不受任何影响。
可以。速度在 0.7× 到 1.3× 之间调节，可以把语速快的参考音慢下来，或者稍微推快一点。画面比例支持 9:16（竖屏，适合小红书、抖音、TikTok、Reels、Shorts）、3:4、1:1。
不需要。所有功能，包括文本转语音和智能剪辑的语音识别，都在 iPhone 本地运行。
通过下面任一方式联系我，我本人会回复。
VoiceCue is an independent project. Built with care for creators who'd rather echo a voice they already own than stare at a teleprompter.
If VoiceCue helped you make a video that sounds like you, that's the whole point.
VoiceCue 是一个独立项目，做给那些不愿盯着提词器念稿的创作者。
如果 VoiceCue 帮你做出了一条听起来就是你的视频，那就是它全部的意义。