How to Generate a Voiceover
Pick the right voice and make your course more engaging and accessible.
Good to know
Voiceover generation uses character limits:
- 200,000 characters per 30 days per account
- Up to 5,000 characters per generation
- Trial accounts: up to 10,000 characters
Each generation uses up your available characters.
iSpring Cloud AI and Suite AI share the same 200,000-character limit. If the account also includes iSpring LMS, its character balance is counted separately and does not affect the Cloud AI or Suite AI limits.
How to add a voiceover
There are two ways to generate a voiceover in a scrollable course.
Way 1. From the selected text
Use this when you already have text on the page. The audio will appear before the selected fragment. You can remove the text and keep only the audio.
- Open a scrollable course and select the text.
- Click on AI > Generate voiceover.
- Edit the text or replace it if needed.
Way 2. From the toolbar
Use this when you want to create new audio content from scratch. For example, you can add an audio introduction or a listening activity that doesn’t exist on the page yet.
- Place the cursor where you want to add audio.
- Click on + > AI > Generate voiceover or open AI tools > Generate voiceover.
- Paste your text.
- Open the voice list and listen to samples using the player icon.
- Click on Generate.
- Listen to the result and rewind if needed.
Choose a different voice, then click on Generate. You can also edit the text and create the voiceover based on the updated version.
Each time you generate a voiceover, the number of available characters decreases.To adjust the speed, move the slider and regenerate the audio.
Speech pace adjustment is supported only in Eleven Multilingual v2. With Eleven v4 Turbo voices, you can adjust the speed using tags such as [slowly], [quickly], etc. A tag applies only to the phrase following it, so to adjust the pace of the entire text, you’ll need to insert multiple tags throughout the text.
- Select the result and click on Insert.
Done! Your voiceover is ready.
You can add a title and description to the file. To do this, click on the voiceover and open its settings. You can also choose the style and color of the text.
How to edit a voiceover
- Click on the audio file and select Edit.
- Update the text or choose a different voice.
- Click on Insert.
How to use Recent and Recommended voices
If certain voices are used often, they appear at the top of the list for quick access.
The Recent section shows the voices that are being used in this scrollable course. If a team is working on the course together, the Recent list is shared across all team members.
ElevenLabs may remove voices from the library.
If this happens, the editor will show a warning that the voice is no longer available. New voiceover cannot be generated with this voice, but existing audio in the course will still work.
To find a suitable voice faster, try the Recommended section. It includes curated ElevenLabs voices that work well for most scenarios.
Voice library
To explore more options, open Voice library. You’ll find over 10,000 voices in different languages. Most names and descriptions are in English because the library is built on user-generated content. You can easily find the right voice or accent using filters.
- Open the Generate voiceover menu.
- Click on Voice library.
Use the Model filter to select the voice model that works best for your current task.
Eleven Multilingual v2 Eleven v4 Turbo Sound Smooth and consistent Expressive, lively, and natural Works best for Сourses and lengthy texts Character voiceovers and dialogues Accent adaptation to the target language Preserves nearly all speech characteristics, including the original accent. Adapts the source voice's accent to the target language. For example, a Spanish voice narrating an English text will sound without a Spanish accent. Pronunciation control Emotion and effect tags are not supported, but you can control pronunciation using punctuation:
- To emphasize a word in a voiceover, try writing it in ALL CAPS.
- To place stress on a specific syllable, place the cursor after the stressed vowel and press Alt+769. Results may vary depending on the selected voice.
- To add pauses, use a hyphen (-) for short pauses or an em dash (—) for longer pauses.
Ensures a more accurate pronunciation of words, stress, and numbers. If you need to correct the stress, place the cursor after the stressed vowel and press Alt+769. Results may vary depending on the selected voice.
It also supports emotion and effect tags, such as [happy], [angry], [whispers], [brief pause], etc. A tag applies only to the phrase following it. You can enter tags in any language.
Adjusting speech pace Supports speech pace adjustment. You can adjust the speech rate to make it suitable for your target audience.
Speech pace adjustment is not supported directly, but you can adjust the speech rate using tags such as [slowly], [quickly], etc. A tag applies only to the phrase following it. To adjust the entire text at once, you’ll need to insert multiple tags throughout the text. - Filter voices by content type: eLearning, role-plays, narration, advertisement, entertainment, social media, or character voice.
You can use filters for gender, age, or language (including accents).
The language filter shows voices trained specifically for a particular language. These voices provide the most natural and accurate voiceover results.
If a language is not listed in the filter, voiceover can still work in that language, but the quality may vary depending on the selected voice.
Preview voices with the player icon.
During the preview, the voice is played back in the language selected in the filter.
If no language is selected, or if a preview in the selected language is not available for that voice, the voice is played back in the interface language. If a preview is also not available in the interface language, the voice will play back in the language it was originally created for.
- Mark favorites with a star.
- Open the Favorites only tab to find them later.
- Click on Select voice.
Done! You’re ready to generate voiceovers with your selected voice.
Why the voice I used is no longer available
The iSpring Voice Library includes ElevenLabs voices.
The ElevenLabs Voice Library consists of user-generated content — ElevenLabs users from all over the world upload voices to it. Authors can hide or remove their voices at any time, making them unavailable in the ElevenLabs Voice Library and, consequently, in the iSpring Voice Library.