Start free
Free tool · no card required

Free Audio Translatortranslate audio in 100+ languages

Upload audio in any of 100+ source languages. Get a transcript plus a translation in 13 target languages, with bilingual SRT subtitles and a parallel-text DOCX. Free for clips up to 3 minutes — no account, no card.

Drop audio or video — get a bilingual transcript

MP3 · WAV · M4A · OGG · FLAC · WEBM · MP4 · MOV · MKV · AVI

Anonymous: up to 3 min/ 50 MB per clip. Sign up free for longer clips and history.

  • 100+ source languages, auto-detected
  • 13 target languages, LLM-translated
  • Files deleted after the response returns

How it works

From a foreign-language clip to aligned bilingual captions in 3 steps.

Drop the file, pick the target language, read both sides on screen. Same workflow whether you’re subtitling a Portuguese podcast in English or studying Mandarin from an interview clip.

  1. 01

    Upload a short clip

    Drag and drop — or click choose. MP3, WAV, M4A, MP4, MOV and 5 more formats accepted. Up to 3 minutes anonymously, longer with a free account.

  2. 02

    Pick the target language

    13 target languages live today. Source language is auto-detected from the first seconds — you don't pick it. The tool figures out you uploaded Portuguese and gives you Japanese back.

  3. 03

    Read side-by-side, export

    Time-aligned segments on screen. Download a bilingual SRT (dual-language subtitles, drop into any video tool) or a parallel-text DOCX (Word document, original left / translated right).

What’s in the box

Two-language output ready to ship — not just a translated paragraph.

Original on top, translation below, exported as a standard subtitle file or a parallel-text Word document.

100+ source languages

Auto-detected from the first seconds — drop in Tagalog, Polish, Hindi, or Quechua and the model picks it up.

13 target languages

English, Spanish, French, German, Italian, Portuguese, Dutch, Polish, Russian, Arabic, Japanese, Korean, Chinese — every pairing supported.

Bilingual SRT + DOCX

Dual-language subtitle file (drop into any video tool) and a parallel-text Word document with original and translation row-by-row.

Aligned timestamps

Each cue carries the original and the translation against the same start/end — no manual re-syncing.

Fast — seconds, not minutes

A 3-minute clip is back in roughly 20 seconds end-to-end. Transcription runs first, translation batches every segment in a single call.

Private by default

Files deleted as soon as the response returns. No training on your audio. TLS in transit, EU/US infrastructure.

More free tools

Bilingual subtitles ready? these pair with translation.

Pull the source video, generate monolingual subtitles instead, or convert formats once you have the SRT.

SRT generator

Audio in, monolingual SRT out — same model, 100+ languages, no translation step.

Interview cleaner

After translation — strip filler words, repetitions, timestamps. Browser-only, no upload.

YouTube downloader

Public YouTube video to MP4 — perfect for translating foreign-language interviews.

SRT to TXT

Strip timestamps for a clean parallel-text doc — same content, no cue indices.

VTT to SRT

WebVTT to SubRip in the browser — for editors that only accept SRT input.

Meeting cost calculator

How much your team wastes per year on multilingual meetings without transcripts.

All tools

14 free tools for every step of the audio-to-text workflow — one hub page.

Common questions

10 questions people ask about audio translation.

01Which source languages can I upload?+
100+ languages — including English, Spanish, Portuguese, French, German, Italian, Dutch, Russian, Ukrainian, Polish, Turkish, Arabic, Hebrew, Hindi, Mandarin, Cantonese, Japanese, Korean, Vietnamese, Thai, Indonesian, Swahili, and a long tail of regional languages. The model auto-detects the source from the first seconds — you don't pick a source language up front. See the full list at /languages.
02Which target languages can I translate into?+
Right now the public tool ships 13 target languages: English, Spanish, French, German, Italian, Portuguese, Dutch, Polish, Russian, Arabic, Japanese, Korean, and Chinese (Simplified). These are the languages with reliable LLM translation quality for spoken-language content. We're expanding the list as we add language-specific QA — drop us a note at [email protected] if you need a target language that isn't here.
03How long can my audio be?+
Anonymous (no account): up to 3 minutes. Both transcription and translation are paid LLM calls per request, so the 3-minute cap keeps the tool free without inviting abuse. For longer files, create a free account — the free plan covers 30 minutes per month at no charge, and Pro ($19/mo) lifts that to 600 minutes plus speaker diarization and AI summaries.
04How accurate is the translation?+
Both stages run on production LLMs. Transcription is 95%+ word accuracy on clean audio in widely spoken languages; translation is roughly on par with what you'd see from a careful first pass by a bilingual reviewer — fine for getting the gist, captions, and search-indexable text, not a substitute for certified legal/medical translation. We never claim 100%. Heavy accents, overlapping speakers, and obscure jargon push accuracy down on both stages.
05Does the tool handle accents?+
Yes, within reason. Standard regional accents (UK / US / Australian English, European / Latin American Spanish, Brazilian / European Portuguese, Cantonese-accented Mandarin, etc.) transcribe cleanly. Very thick or low-resource regional accents and code-switching mid-sentence are still hard — same constraint every cloud ASR has. If quality matters, budget a human review pass.
06What's in the bilingual SRT?+
Each cue carries two lines: the original language on top and the translated language directly below, sharing the same timestamp. The format is standard SubRip — YouTube, VLC, Premiere Pro, DaVinci Resolve, OBS, and any modern player can read it. Drop it onto a video and you get dual-language captions out of the box.
07What's in the parallel-text DOCX?+
A Word document with a two-column layout: original on the left, translated on the right, row-by-row, ordered by timestamp. Useful for language students reviewing audio, content producers laying out subtitle copy, and researchers documenting interviews in the source plus working language. Opens in Microsoft Word, Google Docs, LibreOffice, and Apple Pages.
08Can I use the result commercially?+
Yes. You own everything we produce from files you uploaded — there are no royalty claims, no usage tier, no per-export fees on the output. Standard caveats apply: you need rights to the source audio (don't translate other people's copyrighted broadcasts without permission), and you're responsible for what the translation says, especially if you're publishing it.
09Is my audio private?+
Uploaded files are deleted from temp storage as soon as the response returns. The transcript and translation aren't saved to your account (you don't have one yet — you're anonymous). We never train models on customer content. Audio is processed in EU/US infrastructure with TLS in transit. Full details at /privacy.
10What audio formats are accepted?+
Audio: MP3, WAV, M4A, OGG, FLAC, WEBM. Video: MP4, MOV, MKV, AVI — the audio track gets extracted automatically, no need to pre-convert. Max 50 MB on the anonymous tier (plenty for 3 minutes of compressed audio).

Need to translate longer clips? Free plan covers 30 min / month.

Free account: 30 minutes of transcription + translation a month, files up to 2 GB. Pro at $19/mo lifts that to 600 minutes plus native speaker diarization, AI summaries, meeting bots, and every export format.

Sign up free

Already a customer? Open the dashboard