Free tool · works now
VTT & SRT to Markdown Converter
Convert Zoom, Teams, Google Meet, and Whisper subtitle files into clean, readable Markdown notes with speaker turns.
Runs 100% in your browser. Turn fragmented WebVTT and SRT subtitle files into clean, organized meeting notes with unified speaker blocks.
987 characters · 23 lines
# Meeting Transcript ### Elena Vance [00:00:02] Good morning team, let's start by reviewing the audio pipeline latency from sprint 42. We had three goals: zero dropped buffers, macOS 15 compatibility, and ScreenCaptureKit tests. ### Marcus Chen [00:00:11] On the ScreenCaptureKit loopback, everything passed. Latency is down to 4.2 milliseconds. We tested both Zoom and Google Meet running concurrently without any audio bleed. ### Elena Vance [00:00:19] That is fantastic. What about Whisper model memory consumption on base 8GB machines? ### David Kim [00:00:23] The base model stays around 148 MB of RAM. Turbo v3 spikes to 1.6 GB as expected. Overall CPU usage remained under 12% across Apple Silicon hardware.
The friction of working with raw video subtitles
Video conferencing platforms like Zoom, Microsoft Teams, and Google Meet allow meeting hosts to download recorded sessions alongside automated subtitle tracks. These files are typically exported in WebVTT (.vtt) or SubRip (.srt) formats. However, when you attempt to paste these files into a note-taking application or Markdown vault, you are confronted with an unreadable wall of fragmented text.
Subtitles are formatted for visual television displays, not for human reading. A standard subtitle line displays only five to eight words across a two-second window. A thirty-minute call generates over six hundred individual subtitle blocks, repeating timestamps and breaking single sentences across three or four separate cues.
How this converter transforms subtitle files
This utility operates as an intelligent lexical normalizer. Instead of naively stripping numbers, it parses the underlying temporal architecture of the file:
- Speaker turn aggregation: Identifies consecutive subtitle blocks originating from the same speaker and merges them into cohesive, continuous paragraphs. When Elena Vance speaks for twenty seconds across five subtitle cues, this tool merges them into a single clear block.
- Timestamp normalization: Replaces millisecond-heavy timecodes (
00:14:22.840 --> 00:14:26.110) with clean, human-friendly minute markers ([14:22]). - Audio artifact removal: Automatically strips common transcription sound effect indicators such as
[Applause],[Music], and[Laughter]that clutter meeting notes. - Voice tag stripping: Cleans WebVTT voice formatting tags like
<v John Doe>, extracting the clean speaker label and formatting it as a Markdown heading.
Supported subtitle specifications
This browser engine natively parses the two primary subtitle standards used across the web and enterprise video conferencing:
WebVTT (.vtt)
The Web Video Text Tracks format is the W3C standard for HTML5 video captions. WebVTT files begin with a WEBVTT header and use periods to delimit milliseconds (00:01:23.456). WebVTT supports rich metadata including cue identifiers, voice tags (<v Speaker>), and directional alignment settings.
SubRip (.srt)
SubRip is one of the most widely supported caption formats in video production and open-source media tools. It uses sequential integer cue counters, comma-delimited milliseconds (00:01:23,456), and separate lines for subtitle text.
Integrating meeting transcripts into your vault
Once your transcript is converted to clean Markdown, you can drop it directly into your local notes folder. At Oyma, we believe that meeting records belong in plain text files alongside your project documentation. If you want to eliminate the manual step of downloading and converting subtitle files after every call, the Oyma Mac app captures meeting audio directly and transcribes it on-device using Whisper. Learn how meeting recording works without bots, or explore our Meeting Transcript Cleaner for unformatted chat pastes.
Frequently asked questions
Questions
Does this tool upload my subtitle or video files to an external server?
No. The converter operates entirely within your browser tab. Your subtitle text is parsed using JavaScript in memory and never leaves your device.
What is the main difference between WebVTT and SRT formats?
SubRip (SRT) is an older plain text format using comma-separated millisecond timestamps (00:00:00,000). WebVTT (VTT) is an HTML5 standard that uses dot-separated milliseconds (00:00:00.000) and supports speaker metadata tags like <v Speaker Name>.
Why do raw subtitle files have so many repetitive, tiny lines?
Subtitles are designed for video screens, displaying only 30 to 40 characters at a time for 2-3 seconds. When reading as notes, this creates unreadable fragmentation. This tool merges adjacent cues from the same speaker into cohesive paragraphs.
Can I export both timestamped Markdown and plain prose?
Yes. The tool generates structured Markdown with timestamps and speaker headers by default, and also provides a unified reading prose mode suitable for pasting into daily notes.
How does the desktop app handle meeting transcription?
The desktop app captures live microphone and system audio on your Mac, transcribing it in real time using local Whisper models without needing to export or convert subtitle files manually.
Want automated on-device transcription on Mac?
Record calls without bots and transcribe locally using Whisper models in the Oyma app.