Speechify API 2 min read

Launching the Speechify Cookbook

A small, public repo of runnable Speechify recipes. Pick a folder, drop in your API key, run it. TypeScript and Python today, SDK and native REST side by side.

Docs tell you what an API does. A recipe shows you what it looks like to actually use it. The Speechify Cookbook is the second thing: a public repo of runnable recipes for the Speechify TTS API, in TypeScript and Python. Copy a folder, drop in your API key, run it. Every recipe has both an SDK version and a native REST version, so you can see exactly what is on the wire.

The reference docs at docs.speechify.ai stay the source of truth for parameters and response shapes. The cookbook is for the other question: what does it actually look like to use this for X. Each recipe is self-contained on purpose. You do not have to read three READMEs to run the first one.

What ships today

  • quickstart: Synthesize a sentence to output.mp3. The smallest possible useful program.
  • streaming: Write audio to disk as it generates, instead of waiting for the full payload.
  • SSML controls: Emotion, pitch, rate, pauses, and emphasis. The knobs that make synthetic speech sound like it means something.
  • speech-marks: Word-level timestamps, exported as WebVTT. The recipe for caption sync.
  • voice cloning: Clone a voice from a sample, synthesize with it, and delete it. Full lifecycle so personal voices do not accumulate.

Every recipe has (or will have) both flavors: one using the SDK, and one using native REST. For example, you can compare the TS SDK quickstart with the TS native quickstart to see what the same thing without the SDK looks like. Some teams cannot pull an SDK in. The native flavor exists so they have a copy-pasteable answer for the same problem, and so the SDK is not a black box for anyone wanting to know what is on the wire.

How you run one

git clone https://github.com/SpeechifyInc/speechify-api-cookbook
cd speechify-api-cookbook/recipes/audio/typescript/sdk/quickstart
cp .env.example .env        # paste your SPEECHIFY_API_KEY
pnpm install && pnpm start  # writes output.mp3

Get an API key at console.speechify.ai/api-keys. That is the whole onboarding.

What is next

The audio/ namespace is the start, not the scope. Native flavor will fill in across the matrix. Bash and curl recipes are reserved for native only.

The coverage matrix lives in COVERAGE.md. That is the live source for what exists and what is wanted.

Contributions

PRs are welcome. Templates exist in templates/. The CONTRIBUTING.md and agents/creating-a-recipe.md files cover the shape. If a recipe you want is in COVERAGE.md as ⬜, that is the file to claim.