Voices

Voices: the library and your clones.

GET/v1/voices

List voices (the studio voices and your clones)

listVoicesvoices:readread: 600/min per key

The studio voices first, then your organisation's clones, oldest first.

Query parameters

limitinteger
Page size, 1-200 (default 50).
cursorstring
The nextCursor of the previous page.
kind"builtin" | "custom"
builtin: the studio voices only; custom: your clones only.
gender"female" | "male"

Returns

200 A page of voices. A page: data is a list of Voice, and nextCursor fetches the next page.

Fields of data

idstringrequired
Voice id (vo_…).
pattern `^vo_.*`
namestringrequired
languagestringrequired
Its default language (a catalogue code).
languagesstring[]required
The languages this voice is promised in, as catalogue codes (GET /languages): every catalogue language for a studio voice, the ten marked for duos for a cloned voice. Set the language on the persona or the session, not by choosing another voice.
engine"premier" | "global"required
The voice engine a persona speaking with this voice runs on: global for the studio voices, premier for cloned voices.
genderstring | nullrequired
female or male for a studio voice; null when not known.
kind"builtin" | "custom"required
builtin: a studio voice; custom: your organisation's clone.
createdAtstring (date-time)required
ISO 8601 UTC timestamp.

Errors

invalid_request unauthorized forbidden upgrade_required rate_limited server_error

Request

curl "https://app.zentiqai.com/v1/voices?limit=20" \
  -H "Authorization: Bearer $ZENTIQ_API_KEY"

import { Zentiq } from "@zentiq/node";

const zentiq = new Zentiq({ apiKey: process.env.ZENTIQ_API_KEY });
const page = await zentiq.listVoices({
  limit: 20,
});

import os
from zentiq import Zentiq

zentiq = Zentiq(api_key=os.environ["ZENTIQ_API_KEY"])
page = zentiq.list_voices(limit=20)

Example answer

Response

{
  "data": [
    {
      "id": "vo_cmfq2b7x10000abcd1234",
      "name": "Zoe",
      "language": "en",
      "languages": [
        "en"
      ],
      "engine": "premier",
      "gender": "…"
    }
  ],
  "error": null,
  "requestId": "req_4f1c2e9a7b3d5e6f7a8b9c0d",
  "nextCursor": null
}

POST/v1/voices

Clone a voice

createVoicevoices:writeheavy: 20/min per keyIdempotency-Key

Send 10 to 60 seconds of clean speech as JSON (audioBase64), or as a multipart sample file. consent is required: whose voice it is and that they agreed (in multipart, a JSON string field). Without it the request is refused with 400 consent_required before anything is uploaded. The consent is kept as a consent record the organisation's owners can export and withdraw; withdrawing it retires the voice from new sessions. Cloning is a plan feature (403 upgrade_required).

Body

namestring
Default "My voice".
audioBase64stringrequired
10 to 60 seconds of clean speech: base64, or a data:audio/… URI (25 MB at most).
formatstring
The sample's format when it isn't in the data URI.
consentConsentrequired
Consent to clone a real person's voice or to make an avatar from their picture. Recorded with who confirmed it (the API key), when, a hashed IP address, the exact wording and its version. The organisation's owners can export records and withdraw them; a withdrawn voice or face stops starting new sessions.

Or a multipart/form-data upload:

namestring
samplestring (binary)required
The audio file: 10 to 60 seconds of clean speech, 25 MB at most.
consentstringrequired
The Consent object as a JSON string.

Returns

201 The new voice. data is Voice.

Fields of data

idstringrequired
Voice id (vo_…).
pattern `^vo_.*`
namestringrequired
languagestringrequired
Its default language (a catalogue code).
languagesstring[]required
The languages this voice is promised in, as catalogue codes (GET /languages): every catalogue language for a studio voice, the ten marked for duos for a cloned voice. Set the language on the persona or the session, not by choosing another voice.
engine"premier" | "global"required
The voice engine a persona speaking with this voice runs on: global for the studio voices, premier for cloned voices.
genderstring | nullrequired
female or male for a studio voice; null when not known.
kind"builtin" | "custom"required
builtin: a studio voice; custom: your organisation's clone.
createdAtstring (date-time)required
ISO 8601 UTC timestamp.

Errors

invalid_request consent_required consent_invalid consent_outdated unauthorized forbidden upgrade_required idempotency_conflict idempotency_in_progress too_large rate_limited server_error upstream_error

Events

voice.clone.completed

Request

curl -X POST "https://app.zentiqai.com/v1/voices" \
  -H "Authorization: Bearer $ZENTIQ_API_KEY" \
  -H "Idempotency-Key: $(uuidgen)" \
  -H "Content-Type: application/json" \
  -d '{
  "audioBase64": "…",
  "consent": {
    "attested": true,
    "subject": "myself"
  }
}'

import { Zentiq } from "@zentiq/node";

const zentiq = new Zentiq({ apiKey: process.env.ZENTIQ_API_KEY });
const result = await zentiq.createVoice({
  audioBase64: "…",
  consent: {
    attested: true,
    subject: "myself",
  },
});

import os
from zentiq import Zentiq

zentiq = Zentiq(api_key=os.environ["ZENTIQ_API_KEY"])
result = zentiq.create_voice({
    "audioBase64": "…",
    "consent": {
        "attested": True,
        "subject": "myself",
    },
})

Example answer

Response

{
  "data": {
    "id": "vo_cmfq2b7x10000abcd1234",
    "name": "Zoe",
    "language": "en",
    "languages": [
      "en"
    ],
    "engine": "premier",
    "gender": "…"
  },
  "error": null,
  "requestId": "req_4f1c2e9a7b3d5e6f7a8b9c0d"
}

GET/v1/voices/{id}

Get a voice

getVoicevoices:readread: 600/min per key

Path parameters

idstringrequired
Voice id (vo_…); the id without its prefix is accepted too.

Returns

200 The voice. data is Voice.

Fields of data

idstringrequired
Voice id (vo_…).
pattern `^vo_.*`
namestringrequired
languagestringrequired
Its default language (a catalogue code).
languagesstring[]required
The languages this voice is promised in, as catalogue codes (GET /languages): every catalogue language for a studio voice, the ten marked for duos for a cloned voice. Set the language on the persona or the session, not by choosing another voice.
engine"premier" | "global"required
The voice engine a persona speaking with this voice runs on: global for the studio voices, premier for cloned voices.
genderstring | nullrequired
female or male for a studio voice; null when not known.
kind"builtin" | "custom"required
builtin: a studio voice; custom: your organisation's clone.
createdAtstring (date-time)required
ISO 8601 UTC timestamp.

Errors

invalid_request unauthorized forbidden upgrade_required not_found gone rate_limited server_error

Request

curl "https://app.zentiqai.com/v1/voices/vo_123" \
  -H "Authorization: Bearer $ZENTIQ_API_KEY"

import { Zentiq } from "@zentiq/node";

const zentiq = new Zentiq({ apiKey: process.env.ZENTIQ_API_KEY });
const result = await zentiq.getVoice("vo_123");

import os
from zentiq import Zentiq

zentiq = Zentiq(api_key=os.environ["ZENTIQ_API_KEY"])
result = zentiq.get_voice("vo_123")

Example answer

Response

{
  "data": {
    "id": "vo_cmfq2b7x10000abcd1234",
    "name": "Zoe",
    "language": "en",
    "languages": [
      "en"
    ],
    "engine": "premier",
    "gender": "…"
  },
  "error": null,
  "requestId": "req_4f1c2e9a7b3d5e6f7a8b9c0d"
}

DELETE/v1/voices/{id}

Delete a cloned voice

deleteVoicevoices:writewrite: 120/min per key

The voice stops being offered at once. Personas that speak with it keep it until they are given another voice; its recording is removed once no persona uses it. Studio voices can't be deleted.

Path parameters

idstringrequired
Voice id (vo_…); the id without its prefix is accepted too.

Returns

200 Deleted. data is object.

Fields of data

deletedtruerequired
idstringrequired
Voice id (vo_…).
pattern `^vo_.*`

Errors

invalid_request unauthorized forbidden upgrade_required not_found rate_limited server_error

Request

curl -X DELETE "https://app.zentiqai.com/v1/voices/vo_123" \
  -H "Authorization: Bearer $ZENTIQ_API_KEY"

import { Zentiq } from "@zentiq/node";

const zentiq = new Zentiq({ apiKey: process.env.ZENTIQ_API_KEY });
const result = await zentiq.deleteVoice("vo_123");

import os
from zentiq import Zentiq

zentiq = Zentiq(api_key=os.environ["ZENTIQ_API_KEY"])
result = zentiq.delete_voice("vo_123")

Example answer

Response

{
  "data": {
    "deleted": true,
    "id": "vo_cmfq2b7x10000abcd1234"
  },
  "error": null,
  "requestId": "req_4f1c2e9a7b3d5e6f7a8b9c0d"
}