# Cloud Mode vs On-Device Mode

> The difference between Walkie's two transcription modes, what each one sends, and how to choose.

Cloud Mode sends your audio to be transcribed and cleaned up in the cloud: it is faster, punctuates better, removes filler words, and formats text for the app you are in. On-Device Mode runs the speech model on your own machine: nothing leaves the computer, it works offline, and dictation in it is unlimited and free forever. Both modes are available on every plan, including Free — this is a privacy choice, not a paid upgrade.

*Category: Dictation · Last updated: 2026-09-01*

---

## Side by side

|  | Cloud Mode | On-Device Mode |
| --- | --- | --- |
| Where audio is processed | Walkie's servers | Your machine |
| Needs internet | Yes | No |
| Removes filler words | Yes | No |
| Adds punctuation and capitalisation | Yes | Basic |
| Formats for the app you are in | Yes | No |
| Uses your custom dictionary | Yes | Yes |
| Cost on Free | 2,000 words a week | Unlimited, forever |
| Cost on paid plans | Unlimited | Unlimited |

## What each mode actually sends

### Cloud Mode

Cloud Mode uploads the audio of a single dictation, transcribes it, runs the transcript through a cleanup step, and returns text. The audio and the transcript are processed under Zero Data Retention: they are not stored after the request completes, and nothing is used to train AI models on any tier.

### On-Device Mode

On-Device Mode does not make a network request at all. The speech model runs locally, so the audio never leaves your computer. This is the mode to use for anything you are not permitted to send to a third party.

> **Important:** Cloud Mode may send additional surrounding context — such as which application you are dictating into — to improve formatting. On-Device Mode never sends anything. If you need an absolute guarantee that no content leaves the machine, use On-Device Mode.

## Choosing a mode

Use **On-Device Mode** when:

- You are working with confidential, regulated, or client-privileged material
- You are offline — on a plane, on a train, or behind a restrictive firewall
- Your employer's policy forbids sending audio to third-party services

Use **Cloud Mode** when:

- You want the fastest, cleanest result with no editing afterwards
- You are dictating long-form prose where punctuation matters
- You want text formatted differently in a terminal than in an email

> **Tip:** Most people leave Cloud Mode on and switch to On-Device Mode for specific work. The mode switch is in the main window and takes one click, so it is practical to change it several times a day.

## Choosing an on-device model

On-Device Mode runs a speech model you download inside the app. Larger models are more accurate and slower; smaller models are faster and lighter on memory.

| Model | Languages | Best for |
| --- | --- | --- |
| **Whisper Turbo** | 100+ | The best general default — accurate and reasonably fast |
| **Whisper Small** | 100+ | Older or memory-constrained machines |
| **Whisper Large** | 100+ | Maximum accuracy when speed matters less |
| **Parakeet** | English only | Fastest and most accurate for English-only dictation |
| **Moonshine** | English only | Very low latency on modest hardware |
| **SenseVoice** | Multilingual | Strong on Chinese, Japanese, Korean |
| **Breeze ASR** | Multilingual | Traditional Chinese and English |

> **Note:** Parakeet is English-only. If you dictate in more than one language, choose a Whisper model — Parakeet will transcribe other languages badly rather than failing loudly.

## Language detection

Cloud Mode detects the language you are speaking automatically and returns text in that language — speak Spanish, get Spanish. The cleanup step stays in your language rather than drifting into English.

> **Important:** On-Device Mode's automatic language detection is less reliable, and some accents are misdetected — English spoken with a Malaysian accent has been detected as Malay. If your on-device transcripts come back in the wrong language, set your language explicitly in **Settings → Transcription** instead of leaving it on automatic.

## Frequently asked questions about cloud Mode vs On-Device Mode

### Is On-Device Mode a free tier and Cloud Mode a paid one?

No. Both modes are on every plan. The Free plan limits Cloud Mode to 2,000 words a week, but On-Device Mode dictation is unlimited on Free forever. Paid plans lift the Cloud Mode cap. The free-forever guarantee covers dictation — meetings and Read Aloud have their own limits on Free, whichever mode records them.

### Does On-Device Mode still use my custom dictionary?

Yes. Dictionary corrections are applied locally after transcription, so they work in both modes.

### What happens if Cloud Mode cannot reach the internet?

Walkie falls back to your downloaded on-device model rather than failing. If no model has been downloaded, the dictation is kept on disk and offered back to you for retry.

### Which mode is more accurate?

For raw word accuracy they are close, and on English the local Parakeet model is excellent. Cloud Mode wins on the finished result because of the cleanup step — punctuation, casing, and filler removal.

---

Source: https://trywalkie.com/en/docs/dictation/cloud-vs-on-device
Walkie help center: https://trywalkie.com/en/docs