AI speech-to-text service by OURS
Upload audio once.
Get a structured document you can use right away.
Built for sermons, meetings, and phone calls with a two-stage engine (Whisper + Gemini), from transcription to correction, summary, and record storage.
During open beta, visitors can validate the workflow with the free 10-hour monthly plan and only upgrade to Pro if the workflow proves useful.
OURS builds products that turn spoken work into reusable records
OURS is focused on workflows where transcription alone is not enough. We connect transcription, correction, summarization, and record saving so teams can move from spoken content to practical documents without extra cleanup.
Problem
Meetings, sermons, and calls often stay trapped in raw notes instead of reusable documentation.
Product direction
mallog24 connects transcription, correction, summary, and saved records in one operational flow.
Operating principle
We only publish verifiable information and expand visible claims after they are measured in production.
A workflow users can understand in three steps
mallog24 is designed so visitors understand the flow from upload to final structured document before they ever log in.
1. Upload audio
The browser checks duration first and warns when the free quota would be exceeded.
2. AI transcription + correction
Whisper and Gemini refine speakers, context, and domain terms together.
3. Structured document ready
TXT, DOCX, and saved records continue in one operational flow.
Before voice sentence → After structured output
Before
“Ad budget was 15% over this week, please share the revised plan by next week.”
“After consultation, medication guidance should be twice daily and shared with guardian.”
After
Meeting Summary
- - Agenda: Ad budget overrun (+15%)
- - Decision: Revised plan by next week
- - Action: Reassign follow-up owner/timeline
Plan overview before signup
Free
10 hours / month
Transcription + correction + saved records included
Pro
KRW 8,800 / month (VAT included), Unlimited
Base KRW 8,000 + VAT 10% (KRW 800)
Priority processing for frequent use
A cleaner workspace for voice intelligence
Instead of listing every technical detail, we tuned this section around real workflow steps so users can understand value and start fast.
AI-powered speech recognition for cleaner transcripts from any recording. Convert sermons, lectures, and meetings into accurate text in minutes.
mallog24.comUpload audio file
AI Speech Recognition
A two-stage Whisper + Gemini pipeline delivers high-accuracy transcription, even for low-quality audio.
Smart Text Refinement
Domain dictionaries plus AI refinement preserve medical, church, and general terminology with precision.
Built for Multiple Transcript Types
Optimized prompts for sermons, calls, and meetings improve accuracy by context.
Fast Processing
Long audio files are processed within minutes using auto-splitting and parallel handling.
Korean + English
Supports both Korean and English with language-optimized dictionaries and correction logic.
Flexible Exports
Export as TXT or Word, copy to clipboard, and generate bulletin-ready summaries.
Start now
Upload your audio and experience AI-generated transcripts.
Questions people ask before adopting mallog24
How long are audio files stored?
Raw audio is handled temporarily for transcription and removed after processing. Structured results remain within the account feature scope.
Does it support English?
Yes. Korean and English are both supported, and users can choose the language before upload.
Can it separate multiple speakers?
Meeting and call modes prioritize speaker-aware formatting. Clear audio with less overlap improves quality.
Can users validate before paying?
Yes. The free plan includes up to 10 hours per month so teams can test the workflow before upgrading.