Hotwords
Teach it the words it keeps getting wrong: product names, acronyms, colleagues. Available with the Qwen3-ASR models.
Free · Open source · Runs on your Mac
Talk, and it types.Nothing leaves your Mac.
ThunderTalk is a voice input app for macOS. Press a key in any app, speak, and your words appear at the cursor. Recognition runs on your own machine: no account, no subscription, no cloud.
Free and MIT licensed. Apple Silicon Mac, macOS 15 or later.
01Dictation
One key, any app.
ThunderTalk lives in the menu bar. Wherever your cursor is, whether that is a browser, an editor, Slack or a terminal, the text goes there.
01
Press
Tap your hotkey to start and tap again to stop. Right ⌘ is the default. Prefer to hold it while you talk? Switch the press mode to Hold in Settings.
Right ⌘
02
Speak
A speech model on your Mac's own chip does the listening. Chinese, English, or both in the same sentence.
03
Done
The text is pasted at the cursor and kept in a private, searchable history. Spoken numbers turn into digits, and your hotwords stay spelled your way.
Numbers written as numbers
Inverse text normalisation turns spoken numbers into digits in English and Chinese, and leaves titles such as 《一千零一夜》 alone.
Speak one language, paste another
Direct mode translates as you speak. Review mode transcribes first, then shows the translation so you can Replace it or Keep the original. Over 100 languages through SeamlessM4T v2, which needs about 9 GB of disk and is comfortable on 16 GB of RAM or more.
Downloads that tell the truth
Progress comes from the real byte count, an interrupted download resumes, and Cancel stops right away.
A guided first run
Four short steps: welcome, permissions, a model, and a box to try it in.
History on your disk
Every dictation is saved to a file in ~/.thundertalk. Search it from Home, delete a line, or clear it all.
Updates itself
When a release is published a small prompt appears. One click downloads it, swaps the app and relaunches.
02Studio
New in 1.5Studio turns recordings into text, and text into speech.
Two workbenches: one for the audio you already have, one for the words you would rather hear. Everything runs on your Mac and nothing is uploaded.
Transcribe
Drop in a recording, get a transcript with timestamps.
Audio or video: m4a, mp3, wav, mp4, mov and more. Switch on Multiple speakers for a conversation and every turn is labelled. Click a speaker to rename it. Decoding uses macOS itself, so there is nothing extra to install.
Interactive Transcribe preview
Sample
onboarding-review.m4a
click to choose another
Export
Click a speaker name to rename it. The exports update as you type.
Speak
Three engines, or your own voice.
Three local engines: VoxCPM2 for studio-quality 48 kHz speech, IndexTTS-2.5 for faithful cloning, Kokoro for small and fast. Each downloads once, when you first pick one of its voices. Set the speed, scrub through the result, and save it as WAV or M4A.
Interactive Speak preview. This page plays no sound.
Built-in voices
My voicesClone engine
Built-in VoxCPM2 and IndexTTS voices are designed from a text description, not recorded from real people.
Speed
Language: detect automatically
0:00 / 0:07
Clara (EN) · VoxCPM2 · 1.0×
Preview only: no audio is played and no file is written.
My voices
Clone your voice from 5 to 15 seconds of audio.
Record yourself reading a short passage, or import a clean recording. The reference is kept on your Mac under My voices, ready whenever you want to hear your own voice read something. It is spoken with VoxCPM2 or IndexTTS, whichever you choose. Measured on an M3 Max, cloning a 3.6-second recording of the maintainer: an independent speech recogniser read the result back with 0 to 3.4% error, and speaker similarity was 0.97 to 0.98, with either engine. Only clone voices you have the right to use.
- 01Record or importFive to fifteen seconds of natural speech in a quiet room, or a clean recording of one person.
- 02Check the wordsThe recording is written out on your Mac. A wrong transcript makes the clone worse, so correct it.
- 03Name it and saveIt appears under My voices, next to the built-in ones.
Speech models download once, then everything works offline.
Simulated voice recording. This page never uses your microphone.
Record
0 s
Simulated. This page never touches your microphone.
My voices
Nothing saved yet
03Privacy
Your voice has nowhere to go.
There is no account to sign into and no server to send audio to. Recognition, translation, transcription and speech synthesis all run inside an app on your own Mac. And because the code is open, you do not have to take our word for it.
- Audio
- Recognised on your Mac and never sent anywhere.
- History
- A plain file in
~/.thundertalkon your disk. Search it, delete a line, or clear it. - Network
- Used for two things only: downloading models and checking GitHub for updates.
- Account
- None. No sign-up, no subscription, no usage limit.
- Code
- MIT licensed. Read it, build it yourself, fork it.
04Models
Choose the engine that suits your Mac.
Every model runs on-device. The app reads your hardware and marks the ones that fit, then downloads them with real progress.
ModelRuns onSizeLanguagesGood for
Speech recognition
SenseVoice-SmallCPU · ONNX241 MB5 languagesSmallest and quickest to start. No hotwords.
Qwen3-ASR 0.6B int8CPU · ONNX940 MB52 languagesRuns on every Mac. Supports hotwords.
Qwen3-ASR 0.6B DefaultGPU · MLX~1.9 GB52 languagesThe default on Apple Silicon. Supports hotwords.
Qwen3-ASR 1.7BGPU · MLX~4.7 GB52 languagesHarder accents and noisy audio. Wants 16 GB of RAM.
MOSS-Transcribe-Diarize 0.9B StudioGPU · MLX~1.8 GB50+ languagesSpeaker labels and timestamps. Powers the multi-speaker mode.
Parakeet-TDT 0.6B v3CPU · ONNX640 MB25 European languagesPunctuation and casing built in.
Parakeet-TDT 0.6B v2CPU · ONNX640 MBEnglishFast English dictation on the CPU (about 50× real time on an M3 Max).
Studio
VoxCPM2 8-bit SpeakGPU · MLX3.2 GBChinese, English and moreStudio-quality 48 kHz speech and voice cloning, about real time. Apache-2.0.
IndexTTS-2.5 8-bit SpeakGPU · MLX1.7 + 2.3 GB5 languagesVery faithful voice cloning in Chinese, English, Japanese, Spanish and Arabic, about real time. 1.7 GB model plus a 2.3 GB encoder. bilibili Model Use License.
Kokoro v1.1 82M SpeakCPU · ONNX364 MBChinese, EnglishSmall and fast, about 4× real time. 103 built-in voices (100 Chinese, 3 English). Apache-2.0.
Translation
SeamlessM4T v2 LargePyTorch · MPS / CPU~9 GB100+ languagesSpeech and text translation. Best with 16 GB of RAM or more.
05Compare
How it compares.
Taken from each product's own website on 29 September 2026. Prices change, so follow the links.
| ThunderTalk | Wispr Flow | Superwhisper | VoiceInk | |
|---|---|---|---|---|
| Price | Free | Free up to 2,000 words a week; Pro $15 a month ($12 billed yearly) | Free plan; Pro subscription or lifetime licence | Free trial; paid licence |
| Open source | Yes, MIT | No | No | Yes, GPL-3.0 |
| Where speech is processed | On your Mac | In the cloud (their security FAQ) | On your device or in the cloud, your choice | On your Mac |
| Platforms | macOS 15 or later (Apple Silicon) | Mac, Windows, iOS, Android | Mac, Windows, iOS, Android | macOS 15 or later |
ThunderTalk also transcribes meetings with speaker labels, reads text aloud with three local speech engines and clones your own voice, all on your Mac. We have not checked whether the others offer these, so they are not in the table. Where ThunderTalk falls short: it needs an Apple Silicon Mac with macOS 15, and it is not yet notarised by Apple.
06Questions
Plain answers.
Something not covered here? Open an issue on GitHub and it will get an answer.
Is ThunderTalk free?
Yes. ThunderTalk is MIT licensed and free to use. There is no account, no subscription and no usage limit, and the source code is on GitHub.
Does it work offline?
Yes. Once a model is downloaded, speech recognition, translation, Studio transcription and text to speech all run on your Mac without a connection. The network is only used to download models and to check GitHub for updates.
Which apps can I dictate into?
Any app with a text cursor: browsers, editors, Slack, mail, terminals. ThunderTalk pastes the text where the cursor is, so it asks for Microphone and Accessibility permission the first time you run it.
What is Studio?
Studio is new in 1.5 and replaces the old Lab page. Transcribe turns an audio or video recording into a timestamped transcript, with optional speaker labels, and exports TXT, Markdown, SRT, VTT or JSON. Speak turns text into speech with three local engines, VoxCPM2, IndexTTS-2.5 and Kokoro, using built-in voices or a voice you clone from a 5 to 15 second recording. It all runs locally.
Do I need ffmpeg for Studio?
No. Studio decodes audio and video with the tools that come with macOS, so there is nothing else to install.
Where does a cloned voice go?
Nowhere. The short reference recording and its transcript are stored in ~/.thundertalk/voices on your Mac, and nothing is uploaded. Delete a voice in the app, or delete the folder, to remove it. Please only clone voices you have the right to use.
Does it handle Chinese and mixed Chinese-English?
Yes. Qwen3-ASR handles 52 languages, including Chinese and English mixed in the same sentence, and the app itself is available in English and Chinese.
Which Macs are supported?
An Apple Silicon Mac (M1 or newer) with macOS 15 or later. The download is built for Apple Silicon only, and the bundled speech libraries need macOS 15. Intel Macs are not supported by the release build.
Why does macOS warn me the first time I open it?
ThunderTalk is not notarised by Apple, which would cost $99 a year, so Gatekeeper shows a warning after a browser download. Open System Settings, go to Privacy & Security and click Open Anyway. Updates that arrive through the in-app updater skip this step, although macOS may ask you to grant Accessibility and Microphone access again.
How is it different from Typeless, Wispr Flow or superwhisper?
They are closed-source products with paid plans; Wispr Flow processes speech in the cloud, and Superwhisper lets you choose local or cloud models. ThunderTalk is free, open source and always local. See the comparison for sources.