A voice, and ears.
61 voices to speak with, and transcription to listen.
Transcription is live. Speech is shipping.
What your agent can do with it
- Transcribe a voice note, a call recording or a meeting clip.
- Turn spoken instructions into text your agent can act on.
- Read answers aloud in one of 61 voices.Rolling out
- Pick a voice that suits your product.Rolling out
How it reaches your agent
From your own codeLive
Upload an audio file and get the text back, on the same key and base URL as your chat requests.
SpeechRolling out
Text in, audio out, in the voice you pick. It is shipping, so this row says Shipping until it is live.
Works with Murage, Wayland, Claude Code, Codex, Cursor, Hermes, OpenClaw, Cline, and anything else with an OpenAI or Anthropic endpoint. See the setup guides.
Price
- SpeakRolling outfrom $26 per 1M characters
- Transcribe$0.10 per minute heard
These are plan prices. Pay as you go adds 12%, applied once.
Limits
- Transcription takes an uploaded file of up to 8 MB.
- Each file is billed for at least 10 seconds.
Why not wire it up yourself?
One account instead of several
No separate sign-up, key and billing page for each provider behind an ability.
One key, one bill
Every ability is billed to your Flux account, on one bill.
Nothing new to maintain
No extra SDKs to study or keep current. Your agent talks to one endpoint it already knows.
Questions
Can I send a link to an audio file?
Not yet. Transcription takes an uploaded file.
When can my agent speak?
Speech is shipping. This page will say Live once it is.
Is it on the MCP tools server?
Not yet. Today your agent reaches transcription through a call from your own code.
One key. Everything behind it.
Your agent keeps its brain. Flux gives it the rest, on one key and one bill.