NALscribe FAQ

1. Speak and enunciate clearly into the microphone.

2. Position the microphone close to the person speaking and away from other noise sources.

3. Use an external microphone for best audio capture, such as a Bluetooth headset or desktop microphone, especially for larger distances or if there is a plexiglass barrier.

4. Switch to online speech recognition mode for slightly higher accuracy than offline mode.

1. NALscribe requires permission to access the microphone and allow speech recognition. Go to the device Settings > NALscribe and check that access has been allowed for both Microphone and Speech Recognition.

2. Check that you have a stable internet connection or switch to Offline speech recognition only mode if your device is compatible [see System requirements below].

3. If you are in Offline mode, the first time that you initiate offline speech recognition for a particular language, it will request the language model to be downloaded. This set-up step may take around 10 minutes and requires an internet connection.

NALscribe adds automatic paragraph formatting to the transcription after short pauses, however, you can add punctuation like question marks, commas, and full stops by saying the name of the symbol. e.g. “Hello Mr Smith comma how are you today question mark”.

This may be beneficial when using NALscribe with people who rely heavily on reading the captions to help them understand.

The transcription can be edited by tapping the text on the screen and an on-screen keyboard will appear. Tap and hold to select large portions of text.

Note, NALscribe includes a profanity filter (English), so offensive words are censored.

The following languages are supported for transcription both online and offline: English (Australia), English (UK), English (US), English (Canada), French, Spanish, Arabic, German, Italian, Danish, Japanese, Chinese (Mandarin), Korean, Portuguese, Russian. To change the language, press the Settings button (Single gear with solid fill The following languages are supported for transcription both online and offline: English (Australia), English (UK), English (US), English (Canada), French, Spanish, Arabic, German, Italian, Danish, Japanese, Chinese (Mandarin), Korean, Portuguese, Russian. To change the language, press the Settings button (Single gear with solid fill) and scroll to select the desired spoken language.

When the “Auto erase text for privacy” setting is Off, a dialog box asking whether you would like to Save or Clear the current transcription appears after the recording is stopped. Saved transcriptions are stored locally on the device only. There is no automatic saving of transcriptions.

When the “Auto erase text for privacy” setting is 20s or 60s, no Save option is presented when the recording is stopped as these modes are designed for enhanced privacy.

– To view a list of past saved transcriptions, press the Past Transcriptions button.

– To view a saved transcription, tap the item to open. You can press the microphone button to continue and add to the transcription.

– To edit a saved transcription, open the transcription, tap the text and an on-screen keyboard will appear.

– To delete a saved transcription from the list, swipe your finger left on the transcription you want to delete, and tap the Trash icon.

Translation is available on devices running iOS 15+.

To translate words or phrases from the transcription, tap and select the text you would like to translate and choose Translate from the selection options shown. You can change the translation language, and play the audio translation.

Translation is currently supported between English (US), English (UK), Chinese (Mandarin), French, German, Italian, Japanese, Korean, Portuguese, Russian, Spanish, and Arabic. Translation languages can be downloaded for offline use by going to Change Language > Manage Languages in the Translate pop-up window within the app, or in the device’s Settings > Translate.

In public spaces, we recommend selecting Clear screen for privacy = 20 seconds (Enhanced Privacy mode). NALscribe can remain in listening/recording mode, and will automatically clear all text from the screen after 20 seconds of no speech inactivity.

We prefer to display the iPad in landscape orientation, with a large font size, and ensure the iPad screen is positioned so the customer can easily view it without turning away from the person speaking. We also recommend keeping the iPad connected to a power outlet.

In private appointments, we recommend selecting Clear screen for privacy = Never. For this option, there is no automatic screen clearing and the transcription for the session is available in the scrollable window. When recording is stopped, a dialog box will appear with the options Clear or Save.

Optionally, Offline speech recognition only can be switched on for data security and no audio or text data is sent over the internet. An offline/online indicator is displayed on the main screen to give the user assurance.

For personal use, we recommend selecting Clear screen for privacy = Never. For this option, there is no automatic screen clearing and you can choose to save the session’s for future reference at the end of the recording period.

No audio information is stored on the device or collected by the developer (NAL). The audio recorded from the microphone is used to perform real-time speech recognition and is discarded when recording is stopped.

In online mode, the audio is transmitted over the internet to Apple’s speech recognition server and the text result is returned.

In offline mode, no data is transmitted and speech recognition is carried out on the device for maximum privacy and security. The developer (NAL) does not collect any data from this app.

We recommend the following equipment:

– An iPad running iOS 12.1 or later is required as a minimum requirement. For Offline mode: iOS 13+ is required*; for translation: iOS 15+ is required.

– An iPad stand (with optional security cable ),

– An external microphone, e.g. Bluetooth headset or desktop mic with on/off button. This is not essential but does help improve the accuracy, and allow greater mobility while using the app.

* Devices that support offline mode are iPad (5th generation or later), iPad Air (3rd generation or later), iPad Mini (5th generation or later), iPhone 6s or later.

No. For greater privacy, flexibility, and use in areas where internet connection is weak or not available, NALscribe works well without an internet connection on selected devices* with iOS 13 or later. (Note, please first initiate offline mode while an internet connection is available to download the relevant language model.)

However, using online speech recognition will give the highest speech recognition accuracy.

* Devices that support offline mode are iPad (5th generation) or later, iPad Air (3rd generation) or later, iPad Mini (5th generation or later), iPhone 6s or later.

To help individuals have access to free live captioning wherever they are, with no on-going costs, we have launched NALscribe on the Apple App Store. This is compatible with iPhone, iPad, iPod touch and Mac with M1 chip.

As Microsoft and Google charge on-going usage fees to use their speech-to-text technology in other apps, we are unfortunately unable to provide NALscribe on these platforms.

We would love to hear your feedback! Please send your comments to feedback@nal.gov.au