Meetings & calls
Business teams
Summaries, decisions and action items on top of the transcript — shared while the meeting is still fresh.
Podcast Transcription
Drop your video or audio hereor
MP4, MOV, MP3, WAV, M4A. Up to 2 GB.EnglishFollows source
Subanana transcribes a podcast episode to text in 95+ languages at 98% average accuracy: upload the audio or video file — up to 8 hours and 30 GB — or paste a public YouTube or Instagram link and the media is fetched directly, with no download and no browser extension. The transcript comes back with speakers identified and punctuation and paragraphs restored automatically, and the same job also exports subtitles. Both export to SRT, VTT, TXT, DOCX, XLSX and Markdown.
No card needed. Preview the first 15 minutes free.
How a recording becomes a transcript
Interview recording
M4A · 58:12 · uploaded
Upload a file, paste a link, or record in the browser95+ languages, up to 10 per file on paid plans
Transcript
00:12
We're moving the launch to the first week of June.
00:47
Fine — but the pricing page has to be final by then.
Transcript
TXT · DOCX · XLSX · Markdown
Subtitles
SRT · VTT
Translation
95+ languages
Summary
Key points · action items
Answers
Ask the transcript anything
Why teams hand their recordings to Subanana
Not a feature list — the things that decide whether a transcript is usable without listening again.
A transcript you can read, not decode
- Speakers separatedEvery line carries who said it, identified automatically.
- Punctuation and paragraphsRestored automatically, so the text reads as prose — not as one unbroken wall.
- Tidied textFiller words are cleaned away while the meaning stays untouched.
- Ask the transcriptQuestion the recording in the editor and get answers grounded in what was said.
Accuracy that is engineered, not promised
- The best model per languageModels are benchmarked continuously; each file goes to the top performer for its language.
- Hallucination detectionSuspect output is caught and re-processed by another engine before it reaches you.
- Your terms, spelled your wayPin names, products and jargon in a glossary that applies across your projects.
- Propose-and-confirm fixesAI proofreading suggests corrections for misheard words; nothing changes until you approve.
Who turns speech into text here
The flow is the same — what differs is the deliverable: a transcript, minutes, subtitles or a summary.
Videos & podcasts
Creators
One transcript becomes subtitles, show notes and quotable lines, ready for every platform you publish on.
Interviews
Journalists & researchers
Quotes must be verbatim and attributed to the right speaker — and ready well before the deadline lands.
Lectures
Students & educators
Long recordings arrive summarized and searchable, so revision starts at the point that actually matters.
How to transcribe a podcast
Add the episode, transcribe, export. Nothing is installed and nothing is downloaded to your machine.
- Add the episodeUpload the audio or video file, up to 8 hours and 30 GB — most podcast audio lives as a file or an RSS download rather than on a social platform. If the episode is published as a public YouTube or Instagram post, paste that link instead and the media is fetched directly.
- Choose the language and outputPick the spoken language and whether you want a transcript document, subtitles, or both. A translation can be added in the same job.
- Let the AI process itEach language is routed to the model that benchmarks best for it, with speaker labels, punctuation and paragraphs restored automatically on the transcript.
- Review and exportCorrect anything in the editor, then export SRT, VTT, TXT, DOCX, XLSX, Markdown — transcript and subtitle files come out of the same job.
What works, and what does not
Two ways in: upload the file, or paste a link. The link path only covers media that is genuinely public, which most podcast audio is not — so upload is the usual route.
- Ways in
- Upload the audio or video file, or paste a public link
- Upload limits
- Up to 8 hours and 30 GB per file
- Accepted links
- Public YouTube and Instagram, including Shorts
- Not supported via link
- Private, members-only, age-restricted or region-locked media — upload the file instead
- Output
- Transcript document or subtitle files from the same job
- Languages
- 95+, with optional translation
- Free tier
- First 15 minutes of each file, 3 files a month
Podcast transcription software in 2026
A podcast transcript is the spoken episode written out as text. Podcasters want one for three concrete reasons, and the reason you want it decides which output you should ask for.
The three jobs a transcript does
The first is show notes and SEO. Search engines index text, not audio, so a published transcript is the only part of an episode a search engine can actually read. The second is repurposing: a transcript is the raw material for a blog post, a newsletter, quote cards or a clip list, and it is far faster to skim a transcript for the good thirty seconds than to scrub the audio. The third is accessibility — listeners who are deaf or hard of hearing, and listeners in a quiet office or a loud train, need the text rather than the sound.
Transcript or subtitles?
These are different files and they are not interchangeable. A transcript is a readable document: speakers labelled, sentences punctuated, paragraphs broken. Subtitles are a timed file (SRT or VTT) meant to be displayed against playback, which is why subtitle lines are short and, correctly, carry little punctuation. If you publish your episode as a video — on YouTube, or as an audiogram — you want subtitles. If you are writing show notes, you want the transcript. Subanana produces both from one job, so you do not have to decide up front or pay twice.
What to check before you trust a transcript
Accuracy averages 98%, and almost all of the variation is audio quality rather than the model. Remote interviews recorded on one shared room mic, heavy crosstalk, and strong accents are where errors concentrate. Two things help more than anything else: record each speaker on their own track if you can, and load a glossary of the names, brands and jargon your show repeats — proper nouns are what a transcript gets wrong most often, and they are also what a reader notices first.
Speakers and long episodes
Interview and panel shows need speaker identification, or the transcript reads as one undifferentiated block. Subanana separates speakers automatically in transcript mode, so a two-hour interview arrives already attributed rather than needing a manual pass. Files run up to 8 hours and 30 GB, which covers a long-form episode without splitting it.
Languages and translation
95+ languages are supported, each routed to the model that benchmarks best for that language rather than to one model for everything. A translation can be added to the same job, and a bilingual subtitle file carries the original and the translation together — useful if your show has an audience outside its recording language.
Questions we getFrequently asked questions
Either. Upload the audio or video file — up to 8 hours and 30 GB — or, if the episode is published as a public YouTube or Instagram post, paste that link and the media is fetched directly. Private, members-only or region-locked media cannot be fetched, so those need a file upload.
Yes. Speaker identification runs automatically in transcript mode, so a multi-guest episode arrives attributed rather than as one block of text.
Yes. The same job exports subtitle files as well as transcript documents — SRT, VTT, TXT, DOCX, XLSX, Markdown — so you do not have to run it twice. Subtitles are timed and carry little punctuation by design; the transcript is the punctuated, paragraphed version.
Accuracy averages 98%. Audio quality drives most of the variation, and a glossary of names and terms improves how proper nouns are transcribed.
You can preview the first 15 minutes of each file free, 3 files a month, without a card. Exporting the full result is a paid feature.
No. We don't train AI models on your recordings or transcripts today. If that changes for new individual accounts, we will not use any content until a switch in Settings lets you turn it off. Content in team and enterprise workspaces is not used unless your organisation agrees. Everything is stored encrypted.
Updated 2026-09-14