Take 10 minutes.
Get hours back every week.
Short walkthroughs that turn the platform into your operations team: automated pipelines, instant follow-ups, AI-handled admin. One video, one less thing to do manually.
How to Clone and Manage AI Voices
Your voice library at https://app.trustpager.com/tools/voices — browse platform voices, save your own, generate AI speech, and manage your recordings.
The Voices page at https://app.trustpager.com/content/voices is where you build and manage your workspace's voice library. You can browse platform-provided voices, save the ones you use most, generate AI speech from any text, and access all your saved recordings — all in one place.
Voices power your AI voice agents and any speech generation you do from the platform. Getting your library set up properly means your voice agents always sound the way you want them to.
Your Voices library
The Voices tab shows every voice available to your workspace. You can filter between:
- Recommended — voices provided by the platform, available to all workspaces
- Workspace — voices you've saved to your workspace
- All — both combined
Each voice card has a play button so you can preview it before committing. Once you find one you like, click Save to Workspace to add it to your library. Saved voices appear in every voice picker across the platform — including when you configure voice agents and generate speech.
Setting a default voice
Once a voice is saved to your workspace, you can mark it as your default. The default voice is pre-selected any time TrustPager needs to pick a voice automatically — for example, when generating speech via the API or MCP without specifying a voice explicitly.
To set a default: hover over a saved voice card and click Set Default. Only one voice can be the default at a time. To remove it, click Unset Default on the same card.
Generate AI speech
The Generate tab converts any text to audio using your workspace voices. This is useful for producing voiceovers, announcements, training materials, or anything else that needs spoken audio.
To generate speech:
- Open https://app.trustpager.com/content/voices and switch to the Generate tab
- Select a voice from the dropdown — your workspace voices appear at the top
- Paste or type your text (up to 5,000 characters)
- Optionally choose a transcript format: None, Sentences, Phrases, or Words — this adds timestamped captions to the output
- Click Generate Speech
The estimated credit cost is shown before you generate. A rough guide: around 50 credits per 100 characters. Once generated, you can play the audio directly in the browser, download it as an MP3, and copy the timestamped transcript if you selected one.
Advanced settings
Expand Advanced Settings to fine-tune the voice output:
- Stability — higher values produce a more consistent, predictable tone; lower values introduce more expressive variation
- Similarity — how closely the output matches the original voice sample
- Style — the intensity of stylistic expression in the voice
The defaults (Stability 0.60, Similarity 0.75, Style 0.40) work well for most use cases. Adjust them if you find the voice sounds too flat or too exaggerated.
Recordings library
Every audio file you generate is saved automatically. The Recordings tab lists all of them with duration, file size, and the date they were created. You can search by name, play any recording back, download it, or delete it permanently.
The stats bar at the top shows your total recording count, combined duration, and total storage used — handy for keeping track of what's been produced.
Using voices in voice agents
When you create or edit a voice agent at https://app.trustpager.com/auto/voice-agents, you select a voice from your library for the agent to speak with. Workspace voices you've saved here will appear in the agent's voice picker. If you want your agent to use a specific voice, save it to your workspace first.
Generating speech via API or MCP
If you're using the TrustPager API or MCP, the ai_generate_speech tool produces audio from any text and saves it to your recordings library. Pass the voice name or provider voice ID — these are visible on each voice card. If you omit the voice, TrustPager uses your workspace default.
The recording is then retrievable via list_voices and list_audio_recordings, and the audio URL is returned directly in the response.
Troubleshooting
- Generate tab isn't visible — you need write access to the Voices resource. Ask your workspace admin to check your permissions at https://app.trustpager.com/settings/permissions
- Voice not showing in agent picker — make sure you've clicked Save to Workspace on the voice. Platform voices only appear in agent pickers once they're saved
- Credit cost is higher than expected — the estimate is based on character count. Very short texts have a 50-credit minimum regardless of length
Two ways to
meet FinalPiece.
Talk to Evie right now, or book a real human for a deeper walkthrough. Whichever feels right.
Pick up the phone.
Hear her in 10 seconds.
- RealHer voice passes for human.
- SmartKnows your prices, hours, and trade.
- PatientPolite, attentive, never has a bad day.
Pencil us in.
60 minutes with a real human · Google Meet · Free.