Speech to Text - Offline AI

BasikCode

Contains ads

About this app

Turn any audio into text — entirely on your device. No internet required, no data ever leaves your phone.

Speech to Text Offline uses on-device AI to transcribe voice memos, podcast clips, and meeting recordings shared from any app. You can also record directly from your microphone. Get an accurate transcript in seconds.

LIVE RECORDING
Record audio directly in the app with one tap. No need to use another app first — just hit Record, speak, and transcribe instantly.

SLIM INSTALL + OPTIONAL AI TOOLS
The app installs at about 250 MB and transcribes out of the box in 99 languages. AI Tools (summarize, translate, sentiment analysis) are an optional one-time download (~3.5 GB) you start whenever you want from Settings — and you can free that space anytime from Settings → "AI Tools storage".

AI-POWERED TOOLS (OPTIONAL DOWNLOAD)
Once you download the AI Tools pack, you can:
- Summarize transcripts into key points
- Analyze sentiment (positive, negative, neutral)
- Translate to 15 languages: English, French, Spanish, German, Italian, Chinese, Japanese, Arabic, Portuguese, Korean, Russian, Hindi, Slovenian, and more
- Auto-detect the language of your audio

All AI processing happens on-device. Your conversations stay yours.

MULTILINGUAL
The entire app interface is available in 15 languages and follows your phone's language setting automatically. You can also change the app language manually. Transcription works out of the box in 99 languages.

SMART PERFORMANCE
- Live processing speed indicator shows how fast your device transcribes
- Adaptive ETA estimates remaining time based on your device's performance
- Built-in hardware benchmark to measure your device's transcription speed
- Haptic feedback when tasks complete

FEATURES
- 100% offline — works in airplane mode, at 30,000 feet
- Live microphone recording with instant transcription
- Supports audio up to 10 minutes (auto-trimmed)
- Accepts WAV, MP3, OGG, Opus formats
- Copy, share, or export transcripts
- Export original audio as WAV
- Light, dark, and system theme
- Auto-transcribe on share
- Keep engine ready mode for instant subsequent transcriptions
- Proactive RAM and language compatibility warnings
- Minimalist, clean design

PERMISSIONS
- Internet: used to display ads and to download the optional AI Tools (one-time, ~3.5 GB). Never used to upload your audio or transcripts.
- Microphone: used only for live recording. Audio is processed entirely on-device.
- Notifications: lets the app notify you when transcription finishes while you're in another app.

ADS
The app displays a small banner ad to support development. All ads are family-friendly (PG-rated). Your audio data is never shared with ad networks.

PRIVACY FIRST
Zero analytics on your audio. Zero cloud processing. Your voice data never leaves your device. Period.

What’s new

Smoother UI: in-app messages now sit above the ad banner and the keyboard, the banner itself adapts to your screen, and we hardened startup against a rare background-restriction crash.