HushMemo: Offline AI Notes

HushMemo turns recordings into private, searchable notes using AI that runs on your device.

3 transcription engines. 2 local summary models. 24 AI templates. 7 languages. 5 export formats. No subscription.

Record a meeting, lecture, interview, idea, or personal memo. HushMemo transcribes your audio, organizes the content, generates summaries, and extracts action items—without uploading your recordings to an AI service.

■ Record and Transcribe

- Record voice notes and optionally keep the original audio
- Import audio or video files
- Detect the spoken language automatically or select one manually
- Choose Whisper, SenseVoice, or Zipformer
- Add custom vocabulary for names, terms, and phrases
- Edit transcripts whenever something needs correcting

SenseVoice prioritizes speed for Chinese, English, Japanese, and Korean. Whisper Small supports all seven languages and is better suited to accents, background noise, and challenging recordings. Zipformer provides low-latency transcription for Chinese and English.

■ Turn Speech into Useful Notes

Choose from 24 localized templates for meetings, lectures, interviews, journals, brainstorming, speeches, transcript cleanup, and more.

- Choose how brief or detailed each summary should be
- Generate different summaries from the same transcript
- Extract and review action items
- Ask the built-in local AI questions about a note
- Create custom templates for repeated workflows
- Tag, pin, search, and organize notes
- Attach images and add your own writing

■ Private by Design

After downloading your chosen models, transcription, summaries, action items, and note chat run locally.

Your recordings, transcripts, summaries, notes, templates, vocabulary, chats, images, and tags stay on your device. HushMemo requires no account, does not sync notes to a cloud account, and does not use tracking.

Turn on airplane mode after downloading your models—the core recording and AI features will continue to work. Network access is used only for model downloads and links you choose to open.

■ On-Device Performance

Estimated processing time for 5 minutes of clear audio on iPhone 15:

- SenseVoice: about 20 seconds
- Whisper Small: about 1–1.5 minutes

Speed varies with the model, language, audio quality, device temperature, and available memory.

■ Export Without Lock-In

- Notes and summaries: TXT, Markdown, PDF, or DOCX
- Transcripts: TXT, SRT, PDF, or DOCX
- Original audio: WAV or M4A
- Optional timestamps for transcript exports

■ Honest Pricing

Comparable transcription and AI summarization services can cost $199 a year or more. HushMemo is a one-time purchase.

No subscriptions, no in-app purchases, no ads, and no per-minute transcription or summary credits.

Transcription, summaries, action items, and note chat run on your device—no recurring AI server costs or monthly usage quotas.

Need help with a refund? Open Settings → Request a Refund to view Apple’s official instructions.

■ Supported Languages

English, 简体中文, 日本語, 한국어, Español, Deutsch, and Français—with automatic language detection. The app interface and all 24 built-in templates are available in every supported language.

■ Good to Know

AI models require an initial download and use local storage. Manage or remove them from Settings.

Because your content is stored locally, deleting HushMemo also deletes its local data. Export important notes and recordings first.

HushMemo is made for meetings, lectures, interviews, journals, voice memos, brainstorming, and every moment when speaking is faster than typing.

Questions or feedback? support@hushmemo.app