Voice inputProElite

Voice Input recognizes the task title and its parameters:

• date and time;
• project;
• tags;
• priority;
• reminder.

For example, if you dictate “Team call tomorrow at 3:00 PM in project Work with tag standup high priority remind in 30 minutes”, the app will recognize everything at once: the title, date, time, project, tag, priority, and reminder.

Important! Projects and tags are recognized only if they already exist in the app. New projects and tags are not created.

You can dictate several tags at once — just list them one after another. For example: “Team call tomorrow at 3:00 PM with tags retrospective standup useful” → a task with three tags: “Retrospective”, “Standup”, “Useful”.

Additional commands are also available:

• “to inbox” — the task goes to the Inbox with no project, date, or reminders;
• “someday” — the task will be marked as postponed;
• “no reminder” or “don’t remind” — no reminder will be set.

1. Press and hold the plus (+) icon in the bottom-right corner of the screen.
2. The first time you use voice input, the app will ask for permission to access your microphone. Grant permission.
3. The speech recognition window will appear. Say the task name. The system will automatically recognize your speech and create the task.

If a local model is selected, a screen saying “Loading local model” may briefly appear before recording starts. Wait until the indicator disappears before you start speaking; otherwise, the beginning of the phrase may not be recognized.

On Android devices without Google services, if the local model has not been downloaded yet, you’ll see the “Set up voice input!” prompt instead of the recording window. From there, you can open the settings and download the model. Learn more in the Offline recognition section.

The Recording Screen

Say the task naturally, for example: “Team call tomorrow at 3:00 PM in the Work project with the meeting tag, high priority, remind me 30 minutes before.”

While the app is listening, “Speak...” appears in the center of the screen. Once speech recognition begins, the recognized text appears instead. An animated waveform below reacts to your voice.

Swipe left to cancel creating the task by voice.

If the app can’t recognize what you said, the screen will close after 5 seconds and you’ll see “We didn’t hear you :(”.

What Happens After Recognition

After recognition, the result screen opens. At the top, you’ll see the Recognized block containing your text — the parameters inside it are highlighted.

Below is the task title and the recognized parameters. By default, parameters are removed from the title — only the part that wasn’t a parameter remains.

For example, from “Team call tomorrow at 3:00 PM in project Work with tag standup high priority remind in 30 minutes”, the title will keep only “Team call”, and everything else will be saved as task parameters. You can manage recognition behavior in the settings. See the “Voice Input Settings” section below for details.

If you don’t tap anything, the task will be saved automatically after 5 seconds — the progress is shown on the checkmark button in the bottom right corner. The “Tap to stay” hint means you can tap anywhere on the empty area of the screen to stop the auto-save and remain on the recognition screen.

What You Can Do on the Result Screen

On the result screen you can:

• tap the cross next to a parameter — only that parameter will be removed;
• tap the card with the recognized task or any parameter name — the detailed task page will open for editing, and all parameters will already be filled in from the recognized values;
• tap “Cancel recognition” — all parameters will be reset. Only Inbox and Medium priority will remain.

Voice input settings are located under Settings → Voice input.

At the top of the screen, you’ll see the Voice input method option. Below it, the currently selected method is shown: System voice input or Local model: .... Tap this option to open the model selection screen, where you can choose between your device’s system speech recognition and a downloaded local model for offline use. Learn more in the Offline recognition section.

Recognition of task parameters in voice input depends on the language selected in the app settings (Settings → General settings → Language). Recognition of all parameters, including dates, is supported in Russian and English.

On the same screen, you can control recognition for each parameter: date, reminder, project, tag, and priority. Two switches are available for each parameter:

1. Recognize [parameter] — determines whether this parameter will be recognized during voice input.

   For example, if Recognize tags is turned off and you dictate “Team call tomorrow at 3:00 PM in the Work project with the meeting tag,” the tag won’t be recognized, and the task name will remain “Team call with the meeting tag.” If the option is turned on, the Meeting tag will be added to the task as a separate parameter.

2. Remove [parameter] from task name — available when recognition for that parameter is enabled. It determines whether the recognized parameter remains in the task name.

   For example, if Recognize tags and Remove tag from task name are both turned on, after recognizing “Team call with the meeting tag,” the task name will become “Team call,” while the Meeting tag will be added separately. If Remove tag from task name is turned off, the tag will be added as a parameter but will also remain in the task name: “Team call with the meeting tag.”

Offline recognition lets you dictate tasks without an internet connection. On Android, it also helps when Google services are unavailable or system voice input can’t be used.

To use offline recognition, you only need to download a local speech recognition model once. After that, speech is processed directly on your device, and the audio isn’t sent to a server.

The task parameters that can be recognized—date, project, tags, priority, and reminder—stay the same. Offline recognition only changes how speech is converted into text, not how task parameters are extracted.

How to download and select a model

1. Open Settings → Voice input.
2. Tap Voice input method.
3. On the next screen, download the model you need and select it. You can also keep using system voice input if it’s available.

There is a ? button to the right of the screen title. Tap it to see a short explanation of what local models are for and how to use them.

Downloaded models

The first section shows the voice input methods that are already available:

• System voice input — speech recognition provided by your device. On Android, this is usually Google; on iOS, it uses built-in system features. This option appears only if system speech recognition is available on the device.
• Local models — one row for each downloaded language. Below the language name, you’ll see the file size and the model’s technical name.

The selected method is marked with a green checkmark and the Selected label. To switch to another method, simply tap its row.

Downloaded models have a ⋮ menu on the right:

• if the model isn’t selected — Select and Delete;
• if the model is selected — Delete only;
• the selected system voice input method doesn’t have this menu.

If you delete the currently selected local model, the app will switch back to system voice input if it’s available.

Available device storage

The second section shows how much storage is free on your device and how much space is already used by downloaded models.

Before downloading another model, check this indicator—models can take up tens of megabytes.

Available for download

The third section lists languages whose models haven’t been downloaded yet.

Currently available:

Russian, English, Français, Deutsch, Português, Türkçe, Italiano, українська, Polski, қазақша, Español.

Tap a language row or the download icon on the right to start downloading. While the model is downloading, the icon is replaced with a progress indicator. You can cancel the download at any time.

If something goes wrong, an error icon will appear. Tap it to see the reason and try again.

Before downloading over a mobile network, the app will warn you that the download may use mobile data.

If there isn’t enough storage space, you’ll see Not enough space with a Retry button. If there is no network connection, you’ll see No network. If the downloaded file is corrupted, you’ll see Model file is corrupted and will need to download the model again.

While the model is downloading, progress is also shown in notifications. Once the model is ready, tap the notification to open the model selection screen.

If Google services aren’t available on your device

On some Android devices, such as devices without Google services, system voice input isn’t available. Previously, this meant that voice task creation couldn’t be used. Now you can simply download a local model.

The first time you try to dictate a task without a downloaded model, you’ll see the Set up voice input! window. It explains that you need to download a local model.

Tap Go to settings to open the voice input settings. The Voice input method option will be highlighted so you can go straight to model download.

If system speech recognition is unavailable and no local model has been downloaded, you may also see the Offline recognition unavailable screen inside the voice input window.

Tap Download offline model to open the settings. On Pro and Elite plans, this will usually take you directly to the model selection screen.

The screen closes automatically after 5 seconds or when you tap the back arrow.

Important! A local model is only used to convert speech into text. The voice input limit on the Free plan—10 recognitions—and unlimited voice input on Pro and Elite work the same way as with system speech recognition.

1. In the free version, 10 voice inputs are available; in the Pro and Elite versions, voice input is available without limits.
2. During the first three recognitions, the hint “You can disable recognition in the settings” is shown. Tap it — the Voice Input settings will open.
3. On Android devices, a voice input widget is available — it lets you create tasks by voice right from your home screen without opening the app. For more details, see the Voice Input Widget section.

Requirements for Voice Input to Work

1. On iOS, allow the app to access your microphone. By default, the app uses the system’s built-in speech recognition. If you want to dictate without an internet connection or system speech recognition is unavailable, download a local model in Settings → Voice input → Voice input method.
2. On Android, allow access to the microphone. System speech recognition requires Google services to be installed, and Google voice input must be selected as the default speech input method on your device. If Google services are unavailable, aren’t working correctly, or there is no internet connection, download a local model in Settings → Voice input → Voice input method. Learn more in the Offline recognition section.

If voice input doesn’t start, check the microphone permission and the selected recognition method in the app settings. On Android devices without Google services, the system voice input option may be missing from the list of models—this is normal. Use a local model instead.