> Source: Onevium official documentation
> Article: Voice dictation: recognition, polish, and version differences
> Original URL: https://onevium.com/docs/voice-dictation
> Language: English
> Updated: 2026-09-09
> Applies to: 1.1.22
> Feature status: Released

---

# Voice dictation: recognition, polish, and version differences

Configure local recognition and optional AI polish, dictate a task, and distinguish stable history from beta enhancements.

## Speech becomes text before it becomes a task

Dictation turns speech into editable composer text. The assistant executes the message after you review and send it.

Version 1.1.22 provides local recognition, optional AI polish, and basic dictation history. Start with that workflow; preview-only controls are listed separately below.

## Prepare the microphone and recognition model

1. Open Settings → Voice in the desktop app.
2. Download a model suitable for your main language. When ready, select Use and confirm it is Active.
3. Allow the operating system's microphone permission when first recording. Check system sound-input settings if you need a particular device.
4. Return to the composer and test a short sentence.

The initial model download needs network access. Recognition runs locally. Downloaded does not mean selected. Language groupings and streaming/offline model presentation can vary by version; use the choices shown in your installed build.

## Choose whether to use AI polish

| Setting              | How to choose                                | What it means                                      |
| -------------------- | -------------------------------------------- | -------------------------------------------------- |
| AI transcript polish | Enable for homophone and punctuation cleanup | On by default; disable to retain raw transcription |
| Polish model         | Select a provider model that already works   | Defaults to a fast, low-cost model role            |

Polish sends transcript text to the selected model and can incur model usage. It is separate from offline recognition. If it fails or times out, inspect the retained raw text before sending. Configure the connection in [Providers](https://onevium.com/docs/providers).

## Dictate your first task

1. Open the intended project and conversation, then click the composer microphone. Defaults are **Shift+Command+D** on macOS and **Ctrl+Shift+D** on Windows/Linux. Customized bindings take precedence.
2. Speak a small task:

```text
Read the attached error report.
Summarize reproduction steps and missing information.
Do not modify files yet.
```

3. Stop recording and wait for recognition and any enabled polish.
4. Check the text in the intended composer, correct it, and send.

## Confirm the transcript is correct

Check numbers, paths, proper names, and constraints such as “do not” or “read only.” Confirm the text entered the intended conversation rather than another split composer.

A transcript means input is ready, not that a task executed. After sending, verify the assistant's actual scope and result.

## Inspect basic history and local recordings

Open Settings → Voice → Dictation history to inspect records, and use Open recordings folder to locate audio. The interface around 1.1.22 offers basic recent history; do not assume every stable build has per-record copy, download, or delete controls.

Local recognition does not mean recordings are never retained. Review audio and text before sharing. Do not assume Privacy → Clear Local Data also removes voice recordings.

## Enhanced controls in the preview

These are enhancements in the reviewed **1.1.23-beta.7 development preview**. Stable-version users do not need to reset the app to match missing controls:

| Control                           | Purpose                                                           | How to verify                                   |
| --------------------------------- | ----------------------------------------------------------------- | ----------------------------------------------- |
| Enable voice input                | Off hides the microphone and disables its shortcut                | Check the composer; on by default               |
| Microphone list                   | Authorize access, then choose system default or a detected device | Record a sentence and confirm the device        |
| Copy transcript                   | Copy a record's text                                              | Paste into an empty draft to check              |
| More actions → Download recording | Export the selected audio                                         | Open the exported file                          |
| More actions → Delete recording   | Delete the selected recording record                              | Confirm the target first, then inspect the list |
| History pagination                | Browse more records                                               | Check dates and content across pages            |

There is no visible retention-days setting; do not infer options from another product.

## Common problems

- **Cannot record:** check OS microphone permission and input device; in the beta, also check the master switch.
- **No transcription:** confirm the model is downloaded and active, then test a short recording.
- **Poor recognition:** choose a suitable language model, reduce noise, then consider polish.
- **Polish failure:** check the provider and model; retain and review the raw text.
- **Missing history buttons:** check the installed version; use the recordings folder in stable builds.

## Next steps

Check the [Voice reference](https://onevium.com/docs/settings/reference#voice) and attach task material through the [tools menu](https://onevium.com/docs/tools-menu). Review the complete request before sending it.