Question · Languages and accents
How do you translate a voice note to text?
Transcribe the voice note in the language the speaker used, check the transcript against the recording, then translate the text. Phone recorders can do the first step, and some can translate the finished memo, but translating a wrong transcript carries its mistakes into the English.
Updated October 17, 20263 min read7 cited facts, each traced to its source
Transcribe the voice note in the language it was spoken in, check that transcript against the recording, and only then translate the text. The order matters. A translation made from a wrong transcript carries its mistakes into the English, and the English then looks finished.
Transcribe in the spoken language before you translate
Phone recorders can turn a memo into text, and that step is what makes a translation checkable. On Samsung's Voice Recorder, you select the recording, tap the Transcript assist icon, and tap Transcribe 1. Samsung says that after a memo is transcribed, you can translate it into other languages, share it, or move it to Samsung Notes 2.
Samsung's US page lists 16 languages for Transcript assist, including Japanese and several regional forms of English, Spanish and Portuguese 3. The list does not say whether every device and region supports all 16, so check it on your phone before you rely on it.
Other recorders work differently. Pixel Recorder's Transcribe again option gives a transcript in languages that its Live Transcription does not support, according to a 2023 report from 9to5Google 4. That is useful when your language is missing from the live list, but it applies to the Pixel app and to the version the report describes.
On an iPhone, a university help article says to tap the speech bubble icon in the bottom left of the recording to generate a transcript 5. On a Mac, Apple says the Voice Memos app can recognize and transcribe the speech in audio recordings 6.
The steps for a voice note, in order
- Record the note, or import it into your recorder app.
- Transcribe it, choosing the language the speaker used rather than English.
- Play the note again and check the transcript line by line. Correct any word that does not match what you heard, starting with names, places, numbers and dates.
- Translate the checked transcript into English with a translation app that supports the language.
- Read the English while you listen to the note again, one phrase at a time, and fix any meaning that has shifted.
- Keep the recording and the transcript together, so you can show what was said and what the English means.
Do not skip step 3, since it is where transcription errors get caught. A voice note is short, so checking it takes a few minutes.
Translating a transcript adds a second source of error
A translation is only as good as the text it starts from. A 2026 study on speech translation found that, on the Vietnamese to English subset of FLEURS, translating what a recognizer produced cost between 6.79 and 10.64 BLEU compared with translating clean transcripts 7. That is one language pair and read speech, so it shows how large the effect can be, not how often a given app makes mistakes. The term for this job is speech translation.
It depends on the language, the phone and the privacy
- Your language. If your language is not on the recorder's list, the transcript may come out in the wrong language or not at all. Test a short note first.
- Your phone. The steps above describe Samsung, Pixel and iPhone apps, and menu names change between versions. Check the current steps for your model before you follow them.
- Where the audio goes. Some transcription runs on the phone and some uploads the audio. Read the app's privacy policy before you record anything confidential, and our guide to voice data retention explains what to look for.
- Who is recorded. Some places require everyone in a recording to agree to it. The overview of recording consent laws lists the main rules, and it is not legal advice.
- Stakes. For a legal, medical or professional note, have a qualified person check the translation rather than relying on a phone.
Where to go next
The guide to translating audio and video to English covers the same two steps for longer recordings and computers. For a recording in a language with less training data, see the question on translating Zulu or Xhosa audio. Japanese voice notes have their own accuracy numbers in the question on translating Japanese audio. For a video, the steps and file formats are in the question on translating a video into English subtitles. The guide to transcribing voice memos covers transcripts without translation.
Evidence
Every number on this page, with where it comes from. Open a fact to see the exact words in its source.
- 1.
- 2.
- 3.
- 4.
- 5.
- 6.
- 7.
Go deeper
Related questions
- Can you translate Zulu or Xhosa audio to English?
- How do you translate a video into English subtitles?
- How do you translate an audio file into another language?
- How do you translate Japanese audio to English?
- Can speech to text handle Hinglish, Spanglish and other mixed languages?
- Can Whisper translate audio?
- Can you dictate in two languages at the same time?
- Can you train dictation software to your accent?