Welcome to the Uberduck Text to Speech API!

You're on your way to building something cool! Here's how to get started.

📝 Create an API key and secret

To authenticate to the Uberduck API, you'll need to create an API key and secret. You can create these in the Manage Account page. Make sure you store your secret somewhere safe—you won't be able to access it again!

Consider upgrading

You can use the Uberduck API for free, but your requests will be queued as a cost control measure. For significantly better performance, you can upgrade to one of our paid API plans on our pricing page.

Generate Speech

Once you have your API and secret you can make an HTTP request to /speak. You can authenticate to the API using Basic Auth, with your key as the username and secret as the password.

curl -u $API_KEY:$API_SECRET \
  'https://api.uberduck.ai/speak' \
  --data-raw '{"speech":"This is just a test.","voice":"zwf"}'
  
# Sample output:
# { "uuid": "abcd1234-7f75-441e-b0f1-1aee401134d9" }

Then, using the UUID from the sample output, poll the /speak-status endpoint.

curl -u $API_KEY:$API_SECRET \
  'https://api.uberduck.ai/speak-status?uuid=abcd1234-96e0-45b3-b036-6feac21764e7'
  
# Sample output:
# {
#   "started_at": "2021-07-09T17:07:58.679918",
#   "failed_at": null,
#   "finished_at": "2021-07-09T17:08:10.445369",
#   "path": "https://uberduck-audio-outputs.s3-us-west-2.amazonaws.com/abcd1234-96e0-45b3-b036-6feac21764e7/audio.wav"
# }

When the request has not yet been processed, path and finished_at are null.

List Voices

You can see the set of supported voices on the website, or use the /voices endpoint.

# NOTE: This will output lots of JSON!
curl 'https://api.uberduck.ai/voices?mode=tts-basic' | jq