Models / Speech and music / ElevenLabs

Eleven Turbo v2.5 (text to speech) API

Eleven Turbo v2.5 (text to speech) API: $0.05 per 1,000 characters, prepaid. A failed generation is not charged.

elevenlabs-turbo-2.5Speech and musicper 1,000 characters

Price · per 1,000 characters, USD

SettingsUplinkSwitch
any settings$0.05
Charged when the request is accepted; refunded in full if the generation fails.

Call it · POST /v1/generations

curl
curl https://api.uplinkswitch.com/v1/generations \
  -H "Authorization: Bearer $UPLINK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "elevenlabs-turbo-2.5",
  "input": {
    "text": "Your order has shipped and arrives Thursday."
  }
}'

# then poll until "status" is "succeeded":
curl https://api.uplinkswitch.com/v1/generations/gen_... -H "Authorization: Bearer $UPLINK_API_KEY"

Settings · in input

NameTypeValuesDefault
voicestring—Rachel
stabilitynumber—0.5
similarity_boostnumber—0.75
stylenumber—0
speednumber—1
timestampsboolean—false
previous_textstring——
next_textstring——
language_codestring——
apply_text_normalizationstringauto, on, offauto
text requiredstring——

Questions

How much does the Eleven Turbo v2.5 (text to speech) API cost?
From $0.05 per 1,000 characters. The price depends on the settings in the table and is fixed when the request is accepted.
How do I get a Eleven Turbo v2.5 (text to speech) API key?
Create one in the console at uplinkswitch.ai/console: prepaid credit from $10, no subscription, and the key is live at once. Send it as the bearer token to /v1/generations with model elevenlabs-turbo-2.5; the same key calls every other model in the catalog.
Is there a subscription?
No. You add credit and each request is paid from it. There is no monthly fee and no minimum per request.
What happens if a generation fails?
It is refunded in full, automatically. You pay only for files you receive.
How long are the files kept?
Thirty days, at a URL on api.uplinkswitch.com. Download what you want to keep.
Can I use the same key for other models?
Yes. The same key calls every image, video, speech and 3D model and every chat model in the catalog.