Speech from $0.70 per million characters, your own voice from $6, and ready-made industry packs you can hear first. Record yourself once and your app answers in your voice. 100,000 characters free, no card.
Industry packs are ready-generated recordings you can download and play offline, with a free sample on every one. We have 47 packs across 23 industries today and are building for every industry.
Browse every pack, hear sample lines and get a free 25-phrase sample. Need an industry we have not built yet? Tell us and it moves up the list.
Record yourself, name the voice, and your app answers in it. Customers hear a consistent, recognisable voice instead of a stock one, in every message you send.
Read a paragraph in the browser, upload 10 to 60 seconds, or describe a voice with meters.
Confirm it is your voice to use. It is private to your account.
Send its id as voice in the same API call.
One key, one request, audio back. Then the extras that save you a week.
Create an account and start building. No card, no trial timer. About an hour of speech.
Get a key →Ready-made, quality-checked phrases for support, scheduling, payments, travel and more. Library phrases play in tens of milliseconds.
Listen →An MCP server, an llms.txt and a one-click setup prompt. Your assistant can wire up voice in minutes.
Describe a persona, pick a voice, and call one endpoint per turn. We handle the model, the memory and the speech.
Build an agent →Record, upload or describe a voice, name it and use it in your apps with the same API call. English supported; French and German in beta.
Use your voice →A public status page with real checks every five minutes and daily uptime, plus a changelog of what changed.
See status →Live generation today in English, Spanish, French, Portuguese, Hindi, Japanese and Mandarin Chinese. Written phrase libraries are being built for Arabic, German, Korean, Russian, Italian, Turkish, Polish, Dutch and more, and each gets a voice as we add engines. Non-English results are marked until a native speaker approves them.
Plain HTTP with an x-api-key header. Keep the key on your server.
curl -X POST https://speakvora.com/api/v1/speak \
-H "x-api-key: $SPEAKVORA_KEY" \
-H "content-type: application/json" \
-d '{"text":"Your order is ready for pickup.","voice":"af_heart","lang":"en"}'
# 200 {"url":"https://speakvora.com/voice/v2/af_heart/..../r1.m4a","lang":"en","voice":"af_heart","characters":32,"format":"m4a"}Every phrase is written, spoken, checked by a second pair of ears and only then added. Library phrases play instantly. A bot adds new everyday lines around the clock.
checked and instant
new lines in the checking queue
with live voices