Can I transcribe a recording in two languages? yes, and here is how to get the best result.
Yes, one file can hold two languages: WhipScribe detects the recording's main language automatically from 99, keeps English terms spoken inside it, and takes files up to 10 hours or 5 GB at about 2 minutes per audio hour. Because detection settles on the main language, a recording made of long blocks in each language reads best uploaded as two files; the first 20 minutes are free to test, and a 3-hour bilingual panel is 180 minutes on the $6 pack of 500.
Two languages, one plan
How detection works, and when to cut the file in two.
The main language is found for you
There is no menu to choose from. The recording is listened to, its main language is identified among 99, and the transcript is written in that language. Swedish, Hebrew, Spanish, French, Afrikaans, Tagalog, Korean and Japanese are all detected the same way, with no setting to remember.
Borrowed words are fine
Most bilingual speech is one language carrying words from another: product names, job titles, a phrase in English in the middle of a Spanish sentence. Those English terms inside the speech are handled, so a meeting held in one language with English jargon needs nothing special.
Long blocks: cut at the switch
A conference where the morning is in French and the afternoon in English is really two recordings sharing a file. Detection settles on one main language, so cut at the changeover and upload both halves; each is then detected on its own language and gets its own chapters.
Want it in a third language?
A finished transcript can be translated into any of 99 languages with the same timestamps, so SRT and VTT captions stay aligned with the video. That is often simpler than asking one transcript to carry two languages at once.
Mixed recordings
Guessing settings for a mixed file wastes a run.
✗ Where mixed files go wrong
Habits that make a two-language recording harder than it needs to be.
- Forcing a single language in a menu
- Uploading two hours of each tongue as one
- Splitting at every sentence that switches
- Paying for a full run before testing
- Separate tools for each language
- Captions that drift after translation
✓ What works here
A few choices made once, before the upload.
- Main language detected automatically
- English terms inside the speech kept
- Long single-language blocks sent as two files
- First 20 minutes free to test the mix
- One account for every language
- Translation that keeps the timestamps
Sample output
A mostly Spanish meeting with English terms in it.
Illustrative lines written for this page, not from a real recording, showing English terms kept inside Spanish speech.
Export
One transcript. Five clean formats.
Every pack exports all five — no format is held back for a higher tier.
Plain text
De-ummed paragraphs. Ready to paste.
SRT captions
Timed to the speech. Every video editor reads this.
WebVTT
HTML5 player + YouTube uploads.
Show notes
Formatted with chapters and pull-quotes.
Machine-readable
Timestamped text for your own tools.
Pricing
Honest pricing, no surprises.
A pack from $6 (500 minutes) opens the full transcript. After that, no subscription and no auto-renew — every tier is a one-time minute pack, and credits never expire. Buy once, use the minutes whenever you need them.
Casual pack
$6/500 min
8 hours of audio — $0.72 an hour. One-time — credits never expire.
- 500 minutes of audio
- All export formats
- AI chat over the whole transcript
- Executive summary & key quote
Best value
$12/2,000 min
33 hours of audio — $0.36 an hour. Credits never expire.
- 2,000 minutes of audio
- Summary, chapters and key quotes
- All export formats
- AI chat over the whole transcript
Bulk pack
$24/5,000 min
83 hours of audio — $0.29 an hour. Credits never expire.
- 5,000 minutes of audio
- Summary, chapters and key quotes
- API access for batch pipelines
- All export formats
FAQ
Questions about two-language recordings.
Can I transcribe a recording that uses two languages?
Yes. The main language is detected automatically and English terms inside it are kept. When the recording falls into long stretches of each language, upload the stretches as separate files so each is detected on its own language.
Do I have to tell it which languages are in the file?
No. There is no language setting; detection covers 99 languages and runs on every upload. That is also why the first 20 free minutes are the right way to see how your particular mix comes out.
What happens to short passages in the second language?
The transcript is written on the main language, so a short answer or a quote in the other language can come back less clean than the rest. Check those moments against the timestamps, or cut them out as their own file if they matter.
What about interpreted events?
When an interpreter follows each sentence, the languages alternate too often to split by hand. If the event recorded the floor and the interpretation on separate channels or files, upload the one in the language you need; otherwise send the floor recording and expect the main language to lead.
Does splitting a bilingual file cost more?
No. Minutes are counted on running time, so a 3-hour file and its two 90-minute halves both use 180 minutes, which fits the $6 pack of 500. Ten such panels, 1,800 minutes, fit the $12 pack of 2,000.
What comes back, and where is the audio processed?
A transcript with timestamps in TXT, DOCX, SRT, VTT and JSON, plus a brief, chapters and the names and numbers that were said. Every recording is transcribed on our own servers, and none is ever used for training.
Related
Related pages on languages.
Two languages on one recording? Test the first 20 minutes free.
Upload my recordingOperated by Neugence Technology Pvt. Ltd. · contact@whipscribe.com · Security · Privacy · Terms