How to Translate Discord Voice Messages

Voice notes are everywhere on Discord, and most servers treat them as untranslatable. They are not. This guide covers the three ways to transcribe and translate a Discord voice message with a bot, what it costs (often nothing), and the honest line between voice notes and live voice calls.

Someone drops a 40-second voice note in your multilingual server. Half the channel speaks Spanish, the sender spoke Korean, and the message just sits there: a waveform nobody can read. Text translation bots have been routine on Discord for years, but most communities still believe audio is out of reach. It is not. A voice note is just speech, and speech can be transcribed, and a transcript can be translated, all inside Discord, in seconds, without anyone leaving the channel.

The short version: react to any voice note with a country-flag emoji and Kiki The Translator replies with the spoken words translated into that flag's language, plus a 🔊 button that plays the translation back as audio. That single mechanic works on every tier, in any channel, with zero setup. Everything else in this guide is refinement.

Voice notes are not voice calls (and why that matters) #

First, the distinction most articles blur. A voice note is a recorded audio message: you hold the microphone button in chat, speak, release, and a playable clip posts to the channel. A voice call is live audio in a voice channel, people talking through microphones in real time.

Recorded voice notes can be translated reliably, because the bot receives a complete audio file it can transcribe. Live calls are a different problem: no Discord bot we know of translates live microphone audio dependably, and Kiki does not claim to either. If you have read that "Discord bots cannot translate voice," the writer was talking about live calls. For voice notes, the claim is simply out of date.

Method 1: react with a flag emoji (works everywhere, free) #

The on-demand path, and the one to teach your members first because it needs no configuration at all:

1. Find the voice note in any channel of a server where Kiki is installed.
2. React with a country-flag emoji for the language you want: 🇫🇷 for French, 🇯🇵 for Japanese, 🇮🇳 for Hindi, 🇰🇷 for Korean. Any of the 106 supported languages works.
3. Read or listen. Kiki transcribes the audio, translates the spoken words into the flag's language, and replies under the voice note. The reply carries a 🔊 button that speaks the translation out loud through your normal Discord audio, no voice channel needed.

One related reaction is worth knowing: react 📝 (the memo emoji) to get the plain transcript in the original spoken language, useful when you understand the language but could not make out the audio. Transcript reactions work on every tier.

Method 2: automatic translation in configured channels #

If a channel already runs in-channel translation, voice notes need no reaction at all. Kiki treats a voice note exactly like a typed message: it transcribes the audio, then routes the transcript through whatever translation the channel is configured for.

1. Configure the channel once. Run /set_2_langs language1: language2: in the channel and pick the two languages your community uses. This two-language setup runs on the free tier.
2. Post a voice note. Speak in either language. Kiki detects which one you spoke automatically.
3. Translations appear as replies. One reply per target language, each with its own 🔊 playback button. In a three or four-language channel (paid tiers), every other configured language gets its own translated reply.

In mirrored channels, the setups created with /set_dedicated_lang inside one server or /set_interserver_translation across two servers, the original voice note is forwarded into every linked channel with a translated transcript replied beneath it. Voice forwarding across mirror groups needs the Meow Majesty tier (6 pooled tokens) or higher.

Method 3: ask Kiki in chat #

On servers with AI chat enabled (Feline Finest tier), you can also handle voice notes conversationally: reply to the voice note and ask Kiki to transcribe or translate it in plain language. This is handy for one-off requests in channels with no translation configured, for example "Kiki, translate this voice note to Brazilian Portuguese and summarize it in one line."

What it costs #

Voice note translation follows the same tier rules as text, and the floor is genuinely free:

1. Free (Purrfect Pals). Two-language channels translate voice notes using the server's pooled daily free words (300 per member per day, claimable on the Kiki website). Flag and 📝 reactions work on any voice note.
2. Whisker Whispers ($3/mo pooled). Adds one and three-language channel setups, voice notes included.
3. Meow Majesty ($6/mo pooled). Adds four-language channels and mirror forwarding, where the original voice note travels into every linked channel with its translation.
4. Feline Finest ($9/mo pooled). Adds AI chat handling of voice notes and the 📢 read-aloud reaction for text messages in voice channels.

Because Kiki's subscription is pooled, those monthly figures split across members: nine members contributing one dollar each unlock the top tier for the whole server.

Frequently asked questions #

Can Discord voice messages be translated?

Yes. A translation bot like Kiki transcribes the voice note with speech-to-text, then translates the transcript. The fastest way: react to the voice note with a country-flag emoji (for example a French flag for French) and the bot replies with the translated text plus a speaker button that plays the translation out loud. This works on every server tier and on any voice note in any channel.

Is Discord voice message translation free?

On Kiki, yes. Voice note transcription and translation run on the free Purrfect Pals tier in a two-language channel, drawing from the same pooled 300 free words per member per day that text translation uses. Flag-emoji reactions on voice notes also work on the free tier. Paid tiers add one, three, and four-language channels and mirror forwarding across linked channels.

Can a bot translate live Discord voice calls in real time?

No bot we know of translates live microphone audio in a Discord voice channel reliably, and Kiki does not either. Live calls and recorded voice notes are different problems. Voice NOTES (the recorded audio messages you send in chat) are fully supported: transcribed, translated, and played back. For live calls you would run a dedicated live-captioning tool alongside Discord.

How do I see the transcript of a Discord voice note without translating it?

React to the voice note with the memo emoji (📝) and Kiki replies with the transcript in the original spoken language. This works on every tier, on any voice note, with no setup.

What languages does voice note translation support?

All 106 languages Kiki translates, both as the spoken source language and as the translation target. The source language is detected automatically from the audio.

Related reading: the step-by-step setup guide covers installing and configuring translation from scratch, and the bilingual server playbook covers the channel structure that keeps multilingual communities together.