Turn text into speech
$0.0700/1k chars · $0.0025 minimum/callEleven Flash v2.5
Generate speech audio from text with a selected voice and output format.
ElevenLabs
Text to speech API for AI agents
Scrollport gives AI agents 2 tools to turn text into speech, from ElevenLabs, through one connection. Prices run from $0.0700/1k chars · $0.0025 minimum/call to $0.1400/1k chars · $0.0025 minimum/call, paid per use from one prepaid wallet with no subscription, provider accounts or API keys. Agents use them over MCP, the CLI or the HTTP API.
Updated
Generate speech with Scrollport
Use verified tools with pay-per-use pricing. Provider access is included. Tools that work in your own apps ask you to connect that account.
Connect your first agent to receive $1 trial credit.
Compare Text to speech API tools
The lowest listed price is Eleven Flash v2.5 from ElevenLabs at $0.0700/1k chars · $0.0025 minimum/call. Compare what each tool returns before choosing.
| Tool | Provider | What it returns | Price |
|---|---|---|---|
| Eleven Flash v2.5 | ElevenLabs | Generate speech audio from text with a selected voice and output format. | $0.0700/1k chars · $0.0025 minimum/call |
| Eleven v3 | ElevenLabs | Generate multi-speaker dialogue audio from scripted text with voice and delivery controls. | $0.1400/1k chars · $0.0025 minimum/call |
2 available tools
Questions about Text to speech API
How much do text to speech API tools cost?
Prices run from $0.0700/1k chars · $0.0025 minimum/call to $0.1400/1k chars · $0.0025 minimum/call. You pay per use from one prepaid wallet with no subscription. Your agent sees the exact price before each run, and connecting your first agent adds $1 of trial credit.
What is the best text to speech API for an AI agent?
It depends on what you need back and how much you will run it. Eleven Flash v2.5 (ElevenLabs, $0.0700/1k chars · $0.0025 minimum/call): Generate speech audio from text with a selected voice and output format. Eleven v3 (ElevenLabs, $0.1400/1k chars · $0.0025 minimum/call): Generate multi-speaker dialogue audio from scripted text with voice and delivery controls. Your agent can inspect the closest match and see its exact price before it runs.
Learn more about text-to-speech
Give your agent the final script, listener context and delivery direction. It can then choose a suitable voice and test the lines most likely to expose pronunciation or pacing problems.
Direct the delivery, not just the voice
Describe audience, language, tone, pace and speaking situation. Mark names, numbers, abbreviations and specialist terms that need exact pronunciation, and say which wording must not change.
How can I improve text-to-speech pronunciation?
Provide phonetic guidance or a spoken equivalent for difficult terms, then test those lines in a short sample. Natural punctuation and shorter sentences can also improve pauses and emphasis.
Divide long narration into reviewable sections
Approve a representative passage first, then generate long scripts in logical chunks with consistent voice direction. Check transitions, loudness and pronunciation before joining the final audio.
Example prompt
This example prioritises the lines most likely to need correction.
Example prompt to give your agent
Use Scrollport's Text-to-speech tools to voice [SCRIPT OR ATTACHMENT] for [AUDIENCE AND USE]. Use [LANGUAGE], a [TONE] delivery at [PACE]. Keep the wording exact. Pronounce [TERM] as [GUIDANCE] and test the section containing [DIFFICULT NAMES OR NUMBERS] first. Show me the price before the paid run.