AI Audio-to-Document

Core Features
1. Meeting Recording

Supports two recording sources: MIC and SYSTEM.
Audio files are stored in the app-specific external storage under the Recordings directory, preventing them from being mistakenly listed by general music players.

2. Transcript Processing

Audio recordings can be submitted for speech-to-text transcription.
Long recordings are supported. Based on current real-world testing, when Gemini’s free quota is available, the app has successfully completed transcription workflows for recordings longer than 3 hours.
Transcripts can also be imported from text files (txt / docx).

3. AI Document Generation

Documents can be generated based on document style, content length, and output sections.
Supported document styles include:

Meeting Minutes

Class Notes

Reflection Report

Action List

Decision Log

Executive Summary

Document Export and Sharing

Generated documents are exported as docx files.
After generation, they can be shared directly through the Android share chooser to other apps or cloud storage services.
The app supports sharing audio recordings, transcripts, and generated documents.

4. File Management

Recording management: play/open, rename, delete, transcribe, import audio files.
Transcript management: import, view, rename, delete, export, share.
Document management: view, share, delete.

First-Time Setup

Open the app.

Go to Settings and confirm the AI Provider, Model, and API Key
(recommended default: gemini-3.1-flash-lite-preview).

Select the recording source (MIC or SYSTEM).

The home screen displays a summary of the workflow:
Recording → Transcript → Document, and highlights that, based on current testing, recordings longer than 3 hours can be processed.

Standard Workflow (From Recording to Document)

On the main screen, tap Start Recording, or import an audio/* file from the recording management page.

After recording ends, the system starts generating a transcript. If a transcript already exists, you can directly import a txt / docx file.

Enter Confirm Document Information.

Enter the document name.

Select the document style and content length
(section options have been simplified to system defaults).

Tap Send to AI to generate the content.

In the result window, tap Export Word, and the Android share chooser will open.

You can then:

Share it to other apps

Save it to cloud storage

Other Operations

On the transcript page, you can import txt / docx files.

On the recordings page, you can import audio/* files.

On the document management page, you can directly share existing documents without regenerating them.

Document Styles

Meeting Minutes (STANDARD, exported as DOCX)

Class Notes (ACTION_FOCUSED, exported as DOCX)

Reflection Report (BRIEFING, exported as DOCX)

Action List (ACTION_ITEMS_REPORT, exported as DOCX)

Decision Log (DECISION_LOG, exported as DOCX)

Executive Summary (EXECUTIVE_SUMMARY, exported as DOCX)

Quick Guide to Document Styles

Meeting Minutes: Best for preserving complete meeting content and follow-up tracking.

Class Notes: Best for organizing concepts, key points, and review checklists.

Reflection Report: Best for reflective documents focused on observation + analysis + recommendations.

Action List: Best for turning transcripts into executable task lists (Task / Owner / Deadline).

Decision Log: Best for preserving decision rationale, alternatives considered, and impact scope.

Executive Summary: Best for a dense 1–2 page summary that helps executives make decisions quickly.