Transcribe Spanish Audio to English: How to Do It in Sintesy
You have a recorded lecture in Spanish, a podcast from Latin America, or a call with a client in Madrid or São Paulo. You want the content in English, but you don’t want to replay it three times with a dictionary open.
In Sintesy, the process has two parts: transcribe in the original language and generate the summary in English. The order matters. Skip the first part and the transcript comes out wrong, and a summary built on a wrong transcript is wrong too.
This guide covers the right setting, the step-by-step for files and for YouTube, and what to do when you need the full text in English.
Why transcribe in Spanish before translating
For most people, reading a foreign language is easier than listening to it. In audio you deal with accents, speed and words that run together. On the page, the same sentence is usually much easier to follow.
A transcript moves the problem from your ears to text. You can reread a passage, search for a term, look up a word and quote an exact line.
The English summary does the rest: you know what the recording is about in two minutes and can decide whether the full transcript is worth reading.
The setting that controls everything: Sintesys language
Sintesy works with three languages: Portuguese, English and Spanish. There is no per-file language picker. An account preference decides:
User menu → Settings → Preferences → Sintesys language
This option, described on screen as “Language used to generate your Sintesys”, controls two things at once:
- The language expected in the transcription. The speech recognition engine gets this information before it listens to the audio.
- The language of the summary and of everything generated from it.
So if the language is set to English and you upload audio entirely in Spanish, the recognition engine will look for English where there is none. The result is swapped words and sentences that make no sense.
Don’t confuse it with Interface language, the separate field on the same screen, which only changes menus and buttons.
| Your audio | Sintesys language | What you get |
|---|---|---|
| Mostly English, with some Spanish terms | English | Transcript and summary in English, Spanish terms preserved |
| Mostly Spanish | Español, when uploading | Transcript and summary in Spanish |
| Spanish, but you want the summary in English | Español to upload, English to refresh | Transcript in Spanish and summary in English |
| Portuguese, but you want the summary in English | Português to upload, English to refresh | Transcript in Portuguese and summary in English |
The rule is to choose by the predominant language of the speech. An English meeting where someone quotes a Spanish proverb is still an English meeting.
Step by step: Spanish audio or video file
1. Set Sintesys language to Español
Open Settings → Preferences and change Sintesys language to Español. For Portuguese audio, choose Português. The interface stays in English as long as you leave the other field alone.
2. Upload the file
In the dashboard, click New → Upload file and select the audio or video. Dragging the file onto the screen works too.
Common formats such as MP3, M4A, WAV, OGG, OPUS, MP4 and MOV are supported. On the free plan, the limits are 20 MB per file, 60 minutes per day and 10 Sintesys per day. On Pro, files can be up to 2 GB.
If this is your first upload, the guide on how to transcribe an audio file to text walks through the flow in more detail.
3. Switch Sintesys language back to English
When processing finishes, the Sintesy will have a transcript and summary in Spanish. Now go back to Preferences and change Sintesys language to English.
This step also protects your next upload: if you leave the setting on Español, your next English recording will be transcribed as if it were Spanish.
4. Regenerate the summary, now in English
Open the Sintesy and click the Refresh button (the circular arrows icon in the Sintesy’s action bar).
Since the transcript already exists, Sintesy does not transcribe the audio again. It generates a new summary from the existing text, using the language configured now: English.
The result is the most useful combination for study and work:
- Transcript in Spanish in the Transcription tab, faithful to what was said, for checking quotes and exact terms.
- Summary in English in the Summary tab, for getting the main ideas fast.
Spanish YouTube videos
On YouTube, the flow can be shorter. When a video has captions, Sintesy uses them instead of transcribing the audio. With an English interface, it looks for English captions first, then Portuguese, then Spanish. The summary is written in your Sintesys language.
In other words: with Sintesys language set to English and a video with Spanish captions, just open New → Summarize YouTube videos and paste the link to get the summary in English.
The exception is a video without captions. In that case, Sintesy downloads the audio and transcribes it, so the file rule applies: switch to Español before pasting the link, then switch back to English and use Refresh afterward.
On the free plan, YouTube videos can be up to 15 minutes long. The guide on how to transcribe a YouTube video with AI covers the rest of the flow.
What if I need the full transcript in English?
The summary covers most cases. But sometimes you need the complete text translated, for example to quote an interview or adapt a training session.
For that, use Sintesy’s Chat. Add the Sintesy as a source and ask for something specific:
Translate the transcript from Spanish into English. Keep proper names, acronyms and technical terms as they are, and flag in brackets any passage that seems incomplete.
For long transcripts, ask in parts: “translate the first 10 minutes”, then “continue from where you stopped”. Smaller requests reduce the risk of skipped passages.
It also works for direct questions without translating everything: “What were the speaker’s three main arguments? Answer in English and quote the original Spanish line.” If you’ve never used Chat with sources, the guide on chatting with a PDF shows the same mechanics.
One honest caveat: this is AI translation. For contracts, medical reports or official documents, treat the translation as a draft and have someone who knows the subject review it.
Spanish WhatsApp voice messages
For a short message, WhatsApp itself may be enough. Its built-in voice message transcription uses a language you choose in its chat settings. If that language is set to English and the audio is in Spanish, the transcript fails or comes out wrong, so change it there.
For long audio, or when you want a summary you can come back to later, upload the file to Sintesy following the steps above.
Common mistakes and how to avoid them
Transcript full of English words nobody said. Sintesys language was set to English when you uploaded. Switch to Español and upload the file again. The Refresh button does not fix this, because it only regenerates the summary from the transcript that already exists.
Summary still in Spanish after Refresh. Check that you changed Sintesys language, not Interface language. They are two different fields on the same screen.
English audio that came out in Spanish. You left the preference on Español after your last upload. Switch back to English and upload again.
Strong accent, background noise or people talking over each other. This affects every language. Before blaming the tool, see why a wrong transcript almost always starts with the audio.
Workflow recap
- Preferences → Sintesys language → Español (or Português).
- New → Upload file (or paste the YouTube link, if the video has no captions).
- Preferences → Sintesys language → English.
- Open the Sintesy and click Refresh to generate the summary in English.
- Need the full text translated? Ask in Chat, in parts.
That’s two setting changes and one extra click compared with a normal upload. In return, you keep the original content intact and get an English summary you can read in a few minutes.
Open Sintesy and try it with that Spanish lecture, talk or podcast that’s been sitting on your list.
