Help & Usage Guide

Traditional voice-activated recorders rely solely on volume thresholds — they start recording whenever the sound exceeds a certain level. This means any loud noise, such as snoring, background sounds, or other environmental noise, can trigger recording, even if it’s not what the user wants.

Our AI-driven recording technology goes far beyond simple noise detection. It uses a trained model to analyze the characteristics of sounds, accurately distinguishing between speech, snoring, and ambient noise. This allows the app to record only meaningful audio, such as actual speech or sleep talking, while automatically skipping snoring, background sounds, and silence.

The result is cleaner, shorter, and more relevant recordings — saving time and dramatically improving recording efficiency.

All exported recordings are saved in the public folder: Download/SpeakTrigger_Export. You can access them directly with your file manager or through the app’s share feature.

The app supports automatic sensitivity adjustment, as well as manual control for different environments:
🌳 Outdoor / noisy places: Choose High to capture faint sounds.
🌙 Indoor / nighttime sleep: Choose Low to reduce snoring and background noise when the phone is placed nearby.
🏠 Normal indoor environment: Choose Medium for balanced performance..

When you deny permissions multiple times, Android automatically blocks further permission requests from the app. To enable recording again, please manually grant the required permissions in your device’s settings:
Open Settings → Apps → SpeakTrigger.
Tap Permissions → Allow Microphone (and Notifications, if needed).
Return to the app and try recording again.

The app uses a foreground service to record audio in the background. Android requires showing a notification while a foreground service is running — this is a system rule to protect your privacy.
If you deny notification permission, the app can’t show that required notification, so the recording service isn’t allowed to start.
To fix this, please go to your phone’s settings and allow Notifications for the app.

The AI model identifies multiple types of vocal and emotional sounds, grouped into the following main categories:
🗣️ Speech
Includes normal talking, child speech, conversations, babbling, narration, whispering, and speech synthesizer output.
😊 Emotion
Includes crying, baby crying, laughter, giggling, chuckling, shouting, yelling, and screaming — capturing both joyful and emotional vocal tones.
🎵 Singing
Refers to a cappella vocal expressions such as singing, rapping, chanting, humming, or yodeling — not suitable for recordings with background music.
💨 Other Vocal Sounds
Includes coughing, throat clearing, sneezing, sighing, whistling, panting, gasping.

Snoring is intentionally excluded. Snoring sounds have cyclical characteristics that need to be recorded continuously — some cycles last less than one second, while others can exceed 30 seconds. SpeakTrigger is designed primarily for speech-triggered recording and automatically skips silent intervals, which would remove the pauses between snores and break their natural rhythm.
In fact, this app is able to intelligently skip snoring sounds and focus only on capturing speech or sleep talking, effectively avoiding interference from continuous snoring noise. If you wish to record snoring, please use our other app, SleepAbove, a professional sleep sound tracking app for snoring and sleep talking, powered by the same AI technology.

In Auto Mode, when you tap the record button, stay quiet for about 2 seconds — the app first measures the background noise level to automatically adjust AI sensitivity. After calibration is complete, recording starts automatically. If any sudden noise occurs during the 2-second calibration period, cancel and start again so the app can properly adjust its sensitivity.

In Manual Mode, if your sleeping environment is already very quiet, just select Low Sensitivity. This minimizes snoring interference and effectively skips most snoring sounds.

A few snoring sounds may still be recorded because their acoustic features overlap with speech. If you select High Sensitivity, the chance of capturing these overlapping snores increases.

The app includes multiple safeguards to minimize the risk of losing recordings. For example, if the battery level drops below 5% without charging, available storage falls under 10 MB, or the system restricts background service operation, the app automatically saves all ongoing recordings and safely exits the service. These proactive measures help ensure your recordings are securely preserved even under extreme or unexpected conditions.

To ensure compatibility with more media players, each recording file is limited to a maximum of 1.8 GB. Once this limit is reached, the current file will be automatically saved and marked, and a new recording segment will begin.

Given the current settings — 16 kHz sample rate, 16-bit depth, and mono channel — each recording file is limited to a maximum of 1.8 GB, which corresponds to approximately 16 hours and 47 minutes of continuous audio. In practice, it is extremely rare for someone to speak continuously for such a long period. Importantly, the file size grows only when actual speech or sound is captured; during periods of silence, the app does not significantly increase the file size, effectively waiting for new audio input.

If the device has more available storage, the app will automatically create additional 1.8 GB files as needed. This segmentation allows continuous recording to extend over multiple days, ensuring that long-term audio capture is fully supported as long as the device is plugged in and has sufficient storage capacity.

All audio and analysis data are stored on your device. We do not sell or share your data. AdMob ads only use anonymized, non-personal information for ad delivery.

Yes. SpeakTrigger can run in the background while monitoring. You can stop tracking anytime via the app or notification.

Android’s battery optimization feature can restrict background services to save power. This means if the system thinks the app is “idle,” it may stop background recording, especially when the screen is off or during long sleep sessions.

The app may be forced to stop by the system, causing recordings to be interrupted or lost. To ensure stable background operation, it’s recommended to disable battery optimization for this app.

Open Settings → Battery → Battery Optimization. Find SpeakTrigger. Select “Don’t optimize” or “Allow background activity”.

In rare cases, some devices (like Xiaomi, Huawei, or Samsung) apply additional background limits. If that happens, the app will automatically save the current recording and gracefully stop the service to avoid data loss.

This app is designed for speech-activated recording, which focuses on detecting human voice and automatically skips silence or non-speech sounds.
You can record a cappella (voice-only) singing effectively, but songs with music accompaniment are not recommended, as the background music may be skipped or partially removed.

The app can only detect the subscription for the currently active Google Play account. If the active account has a subscription, the app will show “Subscribed.” Other accounts without a subscription will appear as “Not Subscribed.”

Yes. This is a limitation of Google Play. The app cannot automatically detect subscriptions across multiple accounts. Users must be signed into the Google Play account that owns the subscription to access it in the app.

💡 Tip: If you experience confusion, you can temporarily sign out of accounts without a subscription and sign in with the account that has the active subscription to access it in the app.