Validate
Markovo checks extension, MIME, media signature, account access, file size, and the confirmed Credit ceiling.
Audio to Markdown · Signed-in Beta
Turn recordings into time-coded Markdown.
Upload a meeting, interview, lecture, voice note, or podcast clip. Markovo transcribes speech locally and keeps timestamps beside the words.
MP3, WAV, M4A, AAC, FLAC, OGG, OPUS, and WMA are supported up to 50 MB and 60 minutes. Cost is 1 Credit per started minute and is confirmed before conversion.
Your transcript is ready
Verified output links appear here after conversion.
Readable transcript, traceable timing.
The bundle records detected language, duration, model, time-coded segments, quality evidence, and the actual Credits settled. The production model is bundled in the conversion image and does not fetch weights during a job.
Transcribe
Local speech recognition applies voice activity detection and emits ordered segments with stable timestamps.
Review
Download output.md or the full bundle with metadata, source map, quality report, manifest, and ZIP.
A transcription Beta with explicit boundaries.
audio-to-markdown
Browser, API, CLI, and MCP share one account, Credits balance, History, and bundle.
Clear speech
Low noise, audible voices, and a consistent recording level produce the safest result.
Names and overlap
Proper nouns, accents, music, crosstalk, and speaker diarization can need human review.
Local model
The conversion runtime uses bundled local model files; the standard retention and deletion policy still applies.
Use the same transcript capability from code or an agent.
API
Confirm a ceiling
curl -X POST https://markovo.net/v1/convert \
-H "Authorization: Bearer ${MARKOVO_API_KEY}" \
-F "file=@meeting.mp3" \
-F "capability_id=audio-to-markdown" \
-F "max_credits=60"CLI
Download the bundle
markovo convert meeting.mp3 \
--out runs/meeting \
--max-credits 60MCP
Bound agent work
markovo_convert({
"input_path": "meeting.mp3",
"max_credits": 60
})