Prepare prompts ahead of time so they play instantly

Send up to 200 sentences per call to /v1/prepare, then speak them later without waiting for generation.

POST /v1/prepare gets sentences ready in the background. Preparing is not billed; speaking them later is billed per character as usual. Use it at deploy time for every fixed prompt you will play.

prepare_prompts.py

import os, sys, requests

API = "https://speakvora.com/api/v1"
H = {"x-api-key": os.environ["SPEAKVORA_KEY"]}


def chunks(items, n=200):
    for i in range(0, len(items), n):
        yield items[i:i + n]


prompts = [line.strip() for line in open(sys.argv[1], encoding="utf-8") if line.strip()]
for batch in chunks(prompts):
    r = requests.post(f"{API}/prepare", headers=H, json={"voice": "af_heart", "lang": "en", "texts": batch}, timeout=60)
    r.raise_for_status()
    j = r.json()
    print("ready", j["ready"], "queued", j["queued"], "rejected", j["rejected"])

Use it

python prepare_prompts.py prompts.txt     # one sentence per line, 1 to 500 characters

What the reply means

ready: already in the library or stored; speaking it is quick.
queued: being generated in the background; ready in a few seconds.
rejected: up to 20 examples with the reason (for example digits outside English).

There is a daily cap on prepared texts per account, and a 429 live_rate_limited reply means you reached it. Then run your script again tomorrow or split the work.