Convert MP3 to text

MP3 is the format most recorders, phones, and podcast hosts export by default, so it’s usually the file you’re starting from, whether that’s one interview or a backlog of twenty episodes. Hushscript transcribes it, separates every speaker automatically, and hands back a clean, unwatermarked transcript you can export in 21 formats. There’s no subscription; you pay for the minutes you use, and we’d rather be upfront about which parts are genuinely free and which aren’t than dress up a paid product as a free one.

Convert an MP3 to text in three steps

Drop your MP3, preview the first 30 seconds free, and sign up to get the full transcript. No account is needed just to see the preview. Your first 30 minutes are free: instant when you validate a card (a $1 hold, released right away, never charged), or with your first purchase if you’d rather pay another way. From there, it’s pay-as-you-go against whichever pack you buy.

What’s actually free, and what isn’t

Hushscript doesn’t pretend transcription is free. Turning speech into accurate text runs on top-tier AI, and that costs money to run, so it’s priced in minutes, pay-as-you-go, no subscription. Add a payment method and the first 30 minutes are free. A card’s temporary $1 hold is released immediately, never charged, or another method credits the minutes at your first purchase. Then pay as you go. Speaker labels on every transcript and the browser-based conversion tools are free. After a transcript exists, you can optionally choose Generate Insights: one free run creates the complete revision-bound result, including summaries, actions, decisions, open questions, quotes, chapters, speaker topics, subject clusters, export suggestions, terminology and profanity findings, and cleanup. Review cleanup in Edit and apply only the changes you approve. Packs run $1.99 for 45 minutes, $5.99 for 5 h (300 min), $11.99 for 15 h (900 min), and $19.99 for 30 h (1,800 min), about $0.67 an hour on the largest pack. See pricing for the full table.

Speaker labels and clean exports

Every transcript separates who said what, automatically. It’s free on every job, not a paid tier. Rename “Speaker 1” to a real name in a click. Export as TXT, SRT, VTT, CSV, TSV, Markdown, HTML, RTF, XML, JSON, DOCX, PDF, XLSX, ODS, ODT, ASS, TTML, SBV, LRC, FCPXML, EDL, or a password-protected .husharchive, for 21 formats with no watermark on any of them. The MP3 itself is deleted the moment your transcript is ready, and transcripts are encrypted at rest, so a leak of our storage would expose only unreadable ciphertext.

Clear a backlog of episodes at once

Got more than one file to transcribe, like a season of back episodes, a stack of recorded interviews, or a week of voice memos? Stage them as a batch: drop files into the queue, adjust language or speaker options per file if one needs something different, then hit start and let them run in order. Nothing transcribes the moment you drop it, so you can review the batch before committing. Your balance determines cost; active-job and service safeguards control how much work runs at once.

Great for podcasts, interviews, and voice memos

MP3 is the default export for most recorders and podcast hosts, so drop in an episode, a recorded interview, or a voice memo pulled off your phone and get back a speaker-labeled transcript ready to edit. It works the same whether it’s a single 20-minute conversation or a two-hour panel; the length limit is generous enough that most recordings never come close to it. See podcast transcription and voice recording to text for the specifics of each.

Other formats

Got a WAV, M4A, AAC, or FLAC instead? Convert it to MP3 free in your browser, or transcribe it directly from audio to text. Starting from a video instead? Convert video to MP3 free first, or skip the extra step and transcribe it straight from video to text.