Turn text into speech
Read up to 2,000 characters aloud into a WAV in the media library, for $0.03 a call.
Pick one of six narrator voices (Sulafat, Charon, Kore, Achird, Erinome, Schedar) or pass a character from GET /api/characters to speak in that actor's designed voice. Energy is calm, confident, energetic or hype, and a style line adds direction in plain words. The WAV key goes into a video edit or a post.
Ask in plain words
- "read this script aloud in a calm female voice"
- "text to speech API for my videos"
- "make a voiceover for the launch clip"
From an agent host (MCP)
Server: https://shapelessai.com/mcp (add it to your host).
speak: Read up to 2000 characters aloud into a WAV in the media library, in a preset voice (Sulafat, Charon, Kore, Achird, Erinome, Schedar) or a character's own voice id from characters_list, with an energy and a style line. Returns its media key. SPENDS CREDIT: about $0.03 of engine cost a default call, billed with the plan's margin; budgetUsd caps it (default 2, max 10) and chargedUsd says what it cost. Needs the write scope.
From code (REST)
Bearer API key with the scope each call names (auth, OpenAPI).
POST /api/speech(write scope): Text to speech (tool speak): {text (<= 2000 chars), voice? (Sulafat, Charon, Kore, Achird, Erinome, Schedar), character? (id or slug, reads in its voice), energy? (calm|confident|energetic|hype), style?, name?, budgetUsd?}. Output mediaKey is a WAV.
The body
textstring, required: What to say, exactly as it should be read. At most 2000 characters.voicestring: A roster name or a character voice id. Default Charon.energystring (calm,confident,energetic,hype): How the read moves: calm (measured and reassuring - explainers, anything serious or reflective); confident (the default: brisk, warm, matter-of-fact - product and brand video); energetic (bright and driving - launches, feature reveals, anything with momentum); hype (creator-loud - short-form hooks, reaction and meme-adjacent posts). Default confident.stylestring: Extra direction in plain words, e.g. "smile on the last line". Gemini voices only.namestring: File name stem. Default: the voice name.budgetUsdnumber: The most this call may spend, in USD (default 2, max 10). Held while the tool runs, released after.
From a terminal (CLI)
shapeless tools run speak --input '<json>'(tools show speakprints the schema)
Example
Refusals
| Status | Meaning |
|---|---|
| 400 | The input does not match the tool's schema; the message names the field. |
| 402 | Not enough credit to hold budgetUsd, or the call costs more than budgetUsd. |
| 404 | No such tool on this deployment. |
| 422 | The tool refused this input; nothing was charged. |
| 502 | The tool failed; nothing was charged. |
Where else
More in Media
Generate an image · Generate AI video · Edit a video · Record a product demo · Render a carousel · Render a thumbnail · Make subtitle files · Call any agent tool
