Documentation Index
Fetch the complete documentation index at: https://docs.beyondwords.io/llms.txt
Use this file to discover all available pages before exploring further.
Overview
BeyondWords works with ElevenLabs to bring you best-in-class premade voices, as well as voice cloning and voice generation capabilities. This makes it easy to create hyper-realistic narration that aligns with your brand. Each project can have a default voice for titles, body content, and . You can change the voice associated with any by using the Editor or API.Voice options
- Premade voices: Choose a ready-to-use voice from the ElevenLabs library. Many languages offer a range of accents, genders, and styles. Best for high-quality results with minimal setup.
- Voice cloning: Use Instant or Professional voice cloning to create a voice based on a real speaker. Best for generating audio with recognizable voices or specific regional accents.
- Voice generation: Create a custom voice from a text prompt. Best for creating a unique voice without the need for recordings or approvals.
Voice models
BeyondWords supports three ElevenLabs voice models: v3, Multilingual v2, and Flash v2.5. The model determines how a voice sounds and behaves, shaping aspects like tone, pacing, and overall naturalness. Each model supports different languages and voice options:| Model | Description | Supported languages | Voice options |
|---|---|---|---|
| v3 | Most expressive and dynamic output. Best for engaging, varied narration. | 74 languages | |
| Multilingual v2 | Most stable and even output. Best for clear, consistent long-form narration. | 29 languages | |
| Flash v2.5 | Most accurate and controlled output. Best for replicating specific voices and delivery styles. | 32 languages |
Customization
You can use the following tools to customize speech output:- Voice settings: Shape a voice’s speaking rate, expressiveness, and more
- Pronunciation rules: Define how specific words or phrases should be spoken
- AI preprocessing: Automatically normalize text for more natural speech
- Pauses: Insert breaks to improve pacing and flow
FAQs
How do I choose the right voice?
How do I choose the right voice?
Start by choosing a voice option based on your needs: premade for simplicity, voice generation for flexibility, or voice cloning for a specific voice. (Your options may be limited by the language you’re working in.)Then, choose a voice that fits your content and audience. Consider tone, accent, and how well the voice works for long-form listening. We recommend previewing a few options with your content before deciding.If you need help choosing a voice, please contact our support team.
How do I set a default voice?
How do I set a default voice?
To set the default voice(s) for your project:
- Go to “Content” → “Preferences” → ” Voices” in your project dashboard
- Use the search and/or filtering options to find your preferred voice
- Click “Use voice” alongside any voice to set it as the default for titles, body content, and summaries, or use the dropdown to select a specific option
Does changing the default voice affect existing content?
Does changing the default voice affect existing content?
What is a voice ID and how do I find one?
What is a voice ID and how do I find one?
When using the API, you’ll need to copy the BeyondWords
voice_id of each voice you want to use and include it in your request.To find and copy the voice_id:- Open your project and go to “Content” → “Preferences” → “Voices”
- Click the settings icon alongside your chosen voice
- If applicable, select the ElevenLabs voice model you’d like to use
- Click the “copy” icon alongside the “Voice ID” field
How do I preview available voices?
How do I preview available voices?
To preview available voices, go to “Content” → “Preferences” → “Voices” in your project dashboard and click the colored icon alongside any voice. You can use filters to limit your search.If you’re not yet a BeyondWords customer, request a demo to learn more about our voices.