---
title: "OpenAI-compatible Giggy speech synthesis"
method: POST
path: "/v1/audio/speech"
---

# OpenAI-compatible Giggy speech synthesis

`POST /v1/audio/speech`

## Request

- Base URL: `https://giggy.ai`
- URL: `https://giggy.ai/v1/audio/speech`
- Auth: HTTP bearer

V21 adapter: MP3 uses Fast completed delivery; PCM uses Streaming PCM16 LE, 24 kHz mono. stream_format=sse is supported only for PCM and wraps its native bytes in base64 deltas. No WAV, mu-law, resampling, incremental text input or instructions. Disable client retries. Interrupted delivery requests canonical cancellation.

## Request body

- object
  - `model` 'giggyspeech', required
  - `input` string, required
  - `voice` string, uuid, required
  - `speed` number
  - `response_format` 'pcm' | 'mp3'
  - `stream_format` 'audio' | 'sse'
  - `sample_rate` 24000
  - `instructions` ''

## Response `200`

Native Streaming PCM16 LE at 24 kHz mono, completed Fast MP3, or an SSE envelope over native Streaming PCM.

## Other responses

- `401`: Invalid or revoked Giggy API key
- `402`: Insufficient credits or API budget
- `422`: Invalid or unsupported input
- `429`: Rate or concurrency limit
- `503`: Canonical service unavailable

---

[API](https://skmtc.dev/giggy/apis/giggy-speech-api.md) · [All operations](https://skmtc.dev/giggy/apis/giggy-speech-api/llms.txt) · [OpenAPI document](https://skmtc.dev/giggy/apis/giggy-speech-api/revisions/ee21192fd550?raw)