A transcript that’s a pleasure to read and correct
Echo separates each speaker, names them when it hears their name and keeps every word with its timestamp. You correct it like a document and the audio stays in sync.
Speakers with names
It separates who’s speaking and adds the name if someone introduces themselves or is called by it.
Every word timestamped
Click a word to hear it; double-click to correct it.
Spanish first
Built for Spanish from Spain, and it also transcribes English, Catalan, French and other languages.
Who says what
A transcript without names isn’t much use. Echo looks for them for you.
- It separates the speakers and starts a new paragraph when the speaker changes.
- If someone introduces themselves (“I’m Laura”) or is called by name, Echo adds it or suggests it.
- Change a name once and it changes throughout the transcript and the summary.
- If one person was split in two, “Re-separate speakers” redoes it without transcribing again.
- Recognising your team’s voices in future meetings is optional and stays off unless you turn it on.
Correct it without losing the audio
Text and audio stay together even when you change words or whole paragraphs.
- Uncertain words are underlined so you can check them first.
- Edit a whole paragraph as text: whatever you don’t change keeps its exact timestamp.
- Find and replace across the transcript, and undo and redo with Ctrl+Z and Ctrl+Y.
- Several people can correct at the same time, and comment on or highlight a moment with mentions.
Spelt right from the start
Proper names and acronyms are what goes wrong most in any transcript. That’s what the glossary is for.
- Type the tricky words (names, brands, acronyms) when you upload, or save them in your account’s glossary.
- Choose the language or let Echo detect it.
- You can hide ID numbers, phone numbers, bank accounts, cards and emails from the text and the summary.
Find it and take it with you
What you transcribe doesn’t stay locked inside Echo.
- Search for a word across all your transcripts and filter by who said it, the date, the folder or the tag.
- Download as Word, PDF, text or SRT and VTT subtitles, with or without names, timestamps and summary.
- Translate the transcript into another language, keeping each paragraph’s time and speaker.
- Share a read-only link, with a password or an expiry date if you like.
How to use it
- 1
Upload the recording
In Files, click “Upload files” or “Paste a link”. If you like, choose the language and add the tricky words.
- 2
Carry on with your day
The file appears at the top with its progress and we email you when it’s done.
- 3
Review and correct
Open it, name the speakers in the Speakers tab and double-click anything that needs fixing.
- 4
Download or share it
Click “Download” to choose the format, or “Share” to create a link.
Frequently asked questions
How accurate is it in Spanish?
In our test with Spanish sentences from the public FLEURS dataset it got 97.4 % of words right. What goes wrong most is unusual proper names: the glossary fixes that.
What files can I upload?
MP3, WAV, M4A, MP4 and almost any audio or video, up to 2 GB each and up to 20 at once. Also links from YouTube, Google Drive and other websites.
Which languages does it transcribe?
Spanish and many others, such as English, Catalan or French. You can choose the language when you upload or let Echo detect it.
Does Echo recognise people’s voices?
Only if you turn it on in Settings, because a voiceprint is biometric data. It’s off by default and you can turn it off whenever you like.
Do my recordings leave the European Union?
No. They’re transcribed and stored on Edorteam’s servers in the European Union, without being sent to other companies, and they aren’t used to train models.
Try it with your next meeting.
Create your account in a minute or ask us for a demo for your team.
