To turn a WhatsApp voice note into text, turn on voice message transcripts under Settings, Chats, then long-press the note and tap Transcribe. The text appears under the player, on your phone, for the languages WhatsApp supports. To translate it, copy the transcript into a translator. For anything else, including voice notes in other languages, replies in the other person's language, or a whole inbox of notes, you need a tool that transcribes and translates automatically.
Voice notes are the part of WhatsApp that every translation trick skips. You cannot copy audio into Google Translate, and a note in a language you do not speak is just noise until someone turns it into words. This guide covers what WhatsApp can do on its own, the workarounds on iPhone and Android, and the automatic option for businesses, where every note is transcribed, translated, and can be played back in your language.
WhatsApp's built-in voice message transcripts
WhatsApp can transcribe voice notes on the phone, without sending the audio anywhere. Open Settings, tap Chats, find Voice message transcripts, switch it on and pick a transcript language; the phone downloads a small language pack. From then on, long-press any voice note and choose Transcribe, and the text appears below the player.
- It transcribes in a handful of languages, and only the one you chose in settings. A note in another language comes out as nonsense or not at all.
- It does not translate. You get the words in the language they were spoken.
- It is one note at a time, on one phone. Transcripts are not shared with anyone else who answers the same number.
- Older phones and older app versions do not have it. Update WhatsApp if the option is missing.
The transcript is made on your phone, which is good for privacy, and the reason the language list is short: each language is a pack the phone has to carry.
Translating the transcript
Once a note is text, every message-translation method works on it. Long-press the transcript and copy it, then paste it into Google Translate, the iPhone Translate app, or an AI assistant such as ChatGPT with "translate this to English". On Android, Tap to Translate in the Google Translate app shows the translation over WhatsApp the moment you copy.
It takes four or five taps per note. For a friend's occasional message that is fine. For a day of customer voice notes it is not, and it still leaves you unable to answer with a voice note of your own in their language.
iPhone workarounds
If WhatsApp's transcript does not cover the language you need, the Translate app that comes with iOS can listen. Open it, choose the two languages, tap the microphone, and play the voice note on speaker from WhatsApp (or from a second phone). It transcribes what it hears and shows the translation. Background noise, music and fast speech throw it off, so it is a last resort rather than a routine.
Live Text and the Notes app do not help here: they read text in images, not audio.
Android workarounds
On Android, the Live Transcribe app from Google (free, made for accessibility) listens through the microphone and writes what it hears in dozens of languages. Play the voice note on speaker next to it. Pixel phones also have the Recorder app, which transcribes recordings in a few languages. Copy the result into Google Translate for the translation.
Both are two-step, hands-on processes, and both struggle with the compressed, often noisy audio of a typical voice note.
What an exported chat gives you
If you export a WhatsApp chat (long-press the chat, More, Export chat), the text file lists every voice note as a placeholder such as "audio omitted". Exporting with media adds the audio files to the zip, but no words. So an export is no way to read voice notes either, which also means any chat analysis built on an export misses everything that was said out loud.
Related: Free WhatsApp chat analyzer (shows how many voice notes a chat has)
Automatic transcription and translation for a business
When voice notes come from customers, the question changes from "how do I read this one" to "how do I never miss one". A shared WhatsApp inbox such as Spun answers that with a setting on the chat. Turn on auto-transcribe, pick the language to show the transcript in, and every voice note that arrives is transcribed on the server and translated, ready under the player before you even open the chat, including the ones that arrived while you were away.
You can also hear the note in your language: Spun offers a translated playback, so a Spanish voice note can be listened to in English. And it works the other way. Record a voice note in your language, and the customer can receive it spoken in theirs. The transcripts are searchable, they feed the chat's AI summary, and the whole team sees them on the shared number.
- Transcription in 50+ languages including regional dialects, with 22 Indian languages.
- Runs on the server, so transcripts are waiting for you; nothing depends on your phone.
- Translated playback in your language, and voice replies in the customer's language.
- Per chat: on for the chats that need it, off for the rest. Personal data can be redacted before any text reaches an AI service.
Getting better transcripts
- Ask for shorter notes. Every transcription tool does better on a 20-second note than on a 3-minute one.
- A note recorded in a car or a shop floor will have errors. The transcript is still usually enough to understand the request; listen to the original for exact numbers and names.
- If a customer mixes two languages in one note, automatic language detection handles it better than a fixed transcript language.
- Keep the original. Every good tool shows the transcript beside the audio, not instead of it.