Speech

List voices

Every voice available for text to speech. 145 voices at the last count, filterable by locale, gender and free-text search.

GET/v1/audio/voices

Public — no API key required — and cached for five minutes, so it is safe to call from a settings screen on page load.

curl "https://api.xkiro.com/v1/audio/voices?gender=female&limit=20"

200 OK

{
  "total": 145,
  "returned": 145,
  "voices": [
    {
      "id": "mexican-female",
      "name": "Mexican Female",
      "locale": null,
      "languageKey": null,
      "gender": null,
      "isVip": false,
      "categories": []
    },
    {
      "id": "female-voice-german",
      "name": "Female Voice (German)",
      "locale": null,
      "languageKey": null,
      "gender": null,
      "isVip": false,
      "categories": []
    }
  ]
}

Query parameters#

All optional. Unknown parameters are ignored.

Field	Type	Description
`locale`	string	Exact match on a voice's locale. Many voices carry `null` here, so this narrows harder than it looks.
`languageKey`	string	Exact match on the provider's language key.
`gender`	string	Exact match, lowercase — for example female or male.
`isVip`	boolean	`1` or `true` keeps only premium voices; `0` or `false` keeps only the rest.
`q`	string	Substring search across id, name, locale and language key.
`offset`	integer	Skip this many voices.
`limit`	integer	Return at most this many. `total` is the count before paging; `returned` is how many you actually got.

Unknown parameters are silently ignored

There is no language parameter — sending one returns the entire catalog as if you had filtered nothing. Use locale or languageKey. Check returned against total to confirm a filter actually applied.

Let people hear a voice before they pick one

Generate one short sample per voice with Text to speech and cache the audio. Choosing from a list of names is guesswork; choosing from three seconds of audio is not.

Using a voice#

Pass the id as voice when generating speech. An unknown ID returns 400 rather than silently falling back to a default — a typo cannot ship a product speaking in the wrong voice. (Verified: an unknown ID returns `invalid_request`.)

curl https://api.xkiro.com/v1/audio/speech \
  -H "Authorization: Bearer $XKIRO_API_KEY" \
  -H "Content-Type: application/json" \
  -o out.mp3 \
  -d '{
    "model": "xkiro-voice",
    "voice": "mexican-female",
    "input": "Your order has shipped."
  }'