One workspace. Many languages.

Multilingual Transcription That Works in Every Language, Accent, and Mixed-Language Recording

Upload from YouTube, Dropbox, or Drive, and Cockatoo auto-detects the language and displays right-to-left text correctly.

Drag a file or paste a link.

Choose Detect language, upload your recording, and review the transcript in its original script.

Hablante 100:00

Expertos de la Universidad de Princeton, de Estados Unidos, y de la Uppsala, en Suecia, dieron a conocer que la nueva especie logró evolucionar en tan solo dos generaciones, aun cuando se estimaba que demoraría mucho más, a causa de la reproducción entre un pinzón de Darwin endém…

Actual Cockatoo V4-Pro transcript excerpt · Listen to the original below
Quick answer

Quick Answer

One Tool For Both Halves: Audio To Text, Then Text To Another Language

Cockatoo transcribes the recording, then translates the result. Upload once, get the original transcript and the translated version without switching tools.

Translates Spoken Audio, Not Only Written Documents

Cockatoo listens to the audio, transcribes what was said, and translates it into the language you need. You get text from speech, not document conversion.

Try your own recording

Drop in a file. Start with the words.

3 free transcripts per day. No credit card required.

Speed examples use V4-Fast, up to 120× real-time. The accuracy results below use Cockatoo V4-Pro. Processing time varies with the recording, mode, and service load.

Drag and drop
MP3MP4M4AMOVAACWAVOGGOPUSMPEGWMV
OR

3 free transcripts. No credit card required.

YouTubeDropboxGoogle DriveYour device
The words. The recording. The results.

Accuracy, language by language.

Cockatoo V4-Pro · Updated

Arabic, in its own words.

FLEURS · Read speech · Cockatoo V4-Pro output

على الرغم من تكهّن بعض العلماء بأنه نظراً لوجودها أيضاً في أحواض نهر سرسواتي الجاف الآن، فيجب أن تسمى "حضارة السند وسرسواتي" بدلاً من "حضارة نهر السند"، بينما يطلق عليها البعض "الحضارة الهارابية" نسبة لقرية هارابا حيث تم التنقيب عن أول مواقع تلك الحضارة في العشرينات.

FLEURS

Public benchmark · Read speech

15.2%WER

Internal Realworld

Human-labeled recordings · Accents and noise

36.3%WER

Word error rate (WER) counts substitutions, deletions, and insertions relative to the reference text. Lower is better. These results measure transcription text, not speaker labeling or formatting.

The sample is one FLEURS recording; the figures summarize the full language evaluation. Internal recordings informed provider selection and are an in-sample diagnostic. Results vary with recording quality and dialect.

Read the full accuracy report
Cockatoo V4-Pro · Published language results · Lower is better
LanguageMetricFLEURSInternal Realworld
ArabicWER15.2%36.3%
DanishWER4.3%16.8%
DutchWER5.6%13.9%
FrenchWER7.8%21.0%
GermanWER2.4%11.1%
HebrewWER6.0%22.0%
IndonesianWER4.9%27.7%
ItalianWER6.2%17.4%
JapaneseCER6.5%19.4%
KoreanWER7.6%31.9%
PortugueseWER11.1%14.8%
RussianWER7.8%22.3%
SpanishWER1.0%14.5%
BulgarianWER2.9%22.8%
CatalanWER4.0%12.6%
ChineseCER2.8%31.9%
CroatianWER5.4%19.1%
CzechWER4.6%14.5%
EnglishWER3.9%English (US): 10.9% · English (global): 10.4%
Filipino (Tagalog)WER7.8%21.6%
FinnishWER9.2%24.9%
GreekWER7.4%16.4%
KhmerCER19.7%22.6%
LithuanianWER5.8%11.0%
MalayWER6.6%37.4%
NorwegianWER4.9%28.7%
Persian (Farsi)WER9.6%21.9%
PolishWER5.7%18.9%
RomanianWER8.0%24.3%
SlovakWER4.7%27.6%
SlovenianWER5.4%25.2%
SwahiliWER14.9%42.6%
ThaiCER6.9%23.2%
TurkishWER2.6%16.4%
UkrainianWER2.8%16.1%
VietnameseWER6.9%8.5%
From audio to understanding

Why Teams Choose Cockatoo for Cross-Language Audio

Cockatoo Publishes Tests.

Cockatoo publishes per-language accuracy tests on 36 languages using the standard public benchmark researchers use, plus real-world audio with accents and background noise. See exactly how the engine performs before you upload a file.

What Our Users Appreciate: Accuracy On Difficult Accents And Dialects, Raw Speed, Ease Of Use

Our users appreciate Cockatoo when the audio is hard: thick accents, code-switching, background noise. They also call out how fast the transcript appears and how simple the interface is.

Manual transcription burns budget and time.

Processing In The US And EU, Storage In The US, Access Only With Your Permission

Cockatoo processes files in the US and EU (Finland). Long-term storage is in the US.

Your data is retained while your account is active, and many users treat Cockatoo as a transcript archive. Delete a file or close your account, and the data is removed.

Support engineers can access your files only with your explicit permission.

Support Is Fully Human, Monday To Friday, Inbox Answered Twice A Day

Every support email is answered by a person, not a bot. The inbox is checked twice a day, Monday through Friday. Maximum response time is half a business day.

Most tickets are answered within a few hours. Email hello@cockatoo.com or support@cockatoo.com. Both route to the same inbox.

From audio to understanding

From Upload To Export, In Any Language

Upload
Transcribe
Review
Export

Upload, Get A Transcript In Seconds, Export Where You Need It

Drag your file into the browser. Cockatoo transcribes it in seconds and lets you download the result as TXT, DOCX, SRT, or VTT. No waiting, no setup.

Upload Audio Or Video, Get Editable Transcript With Speakers And Timestamps

Cockatoo accepts audio and video files, identifies speakers, and adds timestamps to every line. The transcript opens in the browser editor so you can correct names or adjust punctuation before export.

In-Browser Editor For Corrections, One-Click AI Summaries, Unlimited AI Notes On Pro

Click into the transcript to fix a word. Click the summary button to condense it into a few paragraphs. Pro users get unlimited access to Canvas, the AI notes editor that turns transcripts into meeting notes, blog drafts, or study guides.

Camille00:02

Bonjour à tous. Aujourd’hui, nous allons préparer les prochaines étapes du projet.

Lucía00:10

Voy a preparar el calendario y compartirlo con el equipo el jueves.

Alex00:18

Thanks. I'll send the meeting summary today so everyone can prepare for our next discussion.

Multilingual meeting · Illustrative transcript
Made for your work

When You Need Accurate Text from Cross-Language Audio

Interview Recordings Become Quotable Transcripts. Lectures Into Searchable Notes.

Upload the interview, get a timestamped transcript, and pull exact quotes for your article. Record the lecture, transcribe it, and search the text for the concept you need to review.

Add Subtitles To Marketing Videos. Make Video Content Accessible And Searchable.

Export the transcript as SRT or VTT and drop it into your video editor. Your marketing video now has subtitles in the original language or translated into another, and search engines can index the spoken content.

Translate PDFs Into English. Localize Slide Decks Without Rebuilding Them.

Upload a PDF in one language, download it in another. The document stays usable: no broken formatting, no manual copy-paste. Translate a PPTX slide deck and the layout survives.

Translate PDF, DOCX, PPTX, XLSX, TXT Between Languages. Document Stays Usable.

Upload a PDF, Word document, PowerPoint deck, Excel spreadsheet, or plain text file. Cockatoo translates the content and returns a file in the same format, with the layout intact. No broken tables, no missing slides.

Evidence you can explore

Tested Accuracy Across 36 Languages, Not a Single Claimed Percentage

Published Accuracy Tests, 36 Languages, FLEURS Benchmark Plus Real-World Audio

Every test includes licensed audio examples so you can listen and compare.

The Most In-Depth Accuracy Evidence in the Space

Cockatoo shows you the word error rate, the test audio, and the transcript produced for each language. No vague percentages with no data.

Check the language-specific results and listen to the original recording.

Explore the evidence
From audio to understanding

Transcribe in Seconds, Edit in the Browser, Export Where You Need It

Upload
Transcribe
Review
Export

Up To 120x Real-Time In Fastest Mode. One-Hour File In About 30 Seconds.

Cockatoo transcribes at up to 120x real-time in the fastest mode. A one-hour recording returns a finished transcript in roughly 30 seconds. V4-Thinking runs at about 20x real-time, still fast enough that you do not wait.

Video Captioning and Subtitle Export

Upload a video file and Cockatoo transcribes the audio, then exports captions and subtitles in the format you need.

Camille00:02

Bonjour à tous. Aujourd’hui, nous allons préparer les prochaines étapes du projet.

Lucía00:10

Voy a preparar el calendario y compartirlo con el equipo el jueves.

Alex00:18

Thanks. I'll send the meeting summary today so everyone can prepare for our next discussion.

Multilingual meeting · Illustrative transcript
From audio to understanding

Try It on Your Own Files, No Credit Card Required

Self-Serve From The First Minute: Upload, Transcript In Seconds, Export

Create an account, upload a file, and get your transcript in seconds. Export it where you need it.

Your workflow, connected

Less busywork. More understanding.

One hour of audio
30 secondsAt up to 120× real-time

Illustrative processing estimate. Upload time, recording, system load, and model affect actual turnaround.

Compare plans
One recording. More possibilities.

Compartamos los próximos pasos con el equipo.

Illustrative translation · Switch between original and translated text.

Paid plans include unlimited V4-Fast transcription. Highest-accuracy modes use a weekly budget shown in Settings. View file and plan limits.

UploadYour audio or video
ProcessingUnited States · Finland (EU)
StorageUnited States
See our security page for details about data handling and account controls.
Trustpilot4.4 / 5124 reviews · Checked October 31, 2026
United StatesEuropean Union
Questions, answered

Common Questions About Multilingual Transcription

Accuracy reportSupported file typesPlan limits

Yes. Paid plans include unlimited V4-Fast transcription with no monthly hour cap. Highest-accuracy modes use a weekly budget shown in Settings. Single users have transcribed over a million hours on one Pro plan.

Your next conversation. In writing.

Start transcribing free