Voices

The workspace's own voices (cloned and designed), favourites, voice design and voice cloning.

Cinara's library voices are listed by GET /v1/speech/options.

Endpoints:

GET /v1/voices
PATCH /v1/voices/{id}
DELETE /v1/voices/{id}
GET /v1/voices/{id}/preview
GET /v1/voices/favourites
PUT /v1/voices/favourites/{voiceId}
DELETE /v1/voices/favourites/{voiceId}
GET /v1/voice-design/options
POST /v1/voice-designs
GET /v1/voice-designs/{id}
GET /v1/voice-designs/{id}/preview
POST /v1/voice-designs/{id}/save
POST /v1/voices/consent
POST /v1/voices
GET /v1/voice-clones/{id}

Two ids per voice

A workspace voice has two ids. Use the right one:

Id	Looks like	Use it for
`id`	UUID	Managing the voice: `PATCH /v1/voices/{id}`, `DELETE /v1/voices/{id}`, `GET /v1/voices/{id}/preview`, and singing voices.
`providerVoiceId`	opaque string	Speaking with the voice: `voiceId` in speech, dialogue, the compatible text-to-speech path, voice changer, audiobooks, talking avatars, dub revisions, voice previews and favourites.

providerVoiceId is an opaque string. Don't parse it.

The voice object

json

{
  "id": "5a0e3c7b-91d2-4f68-b3a4-6c2e8d1f9b05",
  "source": "cloned",
  "providerVoiceId": "vc_7f2a9c41e8b3",
  "name": "Priya narrator",
  "description": "Warm, even delivery for lessons",
  "gender": "female",
  "prompt": null,
  "createdAt": "2026-09-15T09:12:40.118204+00:00",
  "consentVerifiedAt": "2026-09-15T09:14:02.551930+00:00"
}

Field	Type	Description
`id`	string	The voice's UUID.
`source`	string	`cloned` or `designed`.
`providerVoiceId`	string	The id to speak with.
`name`	string	Up to 40 characters.
`description`	string or null	Up to 120 characters.
`gender`	string or null	`female`, `male` or null.
`prompt`	string or null	For a designed voice, the description it was designed from.
`createdAt`	string	When it was created.
`consentVerifiedAt`	string or null	For a cloned voice, when its consent statement was verified.

List workspace voices

GET /v1/voices

Lists the workspace's cloned and designed voices, newest first.

Query parameters

Name	Required	Description
`source`	no	`cloned` or `designed` to list only one kind. Anything else lists both.

Response

200 OK

json

{
  "voices": [
    {
      "id": "5a0e3c7b-91d2-4f68-b3a4-6c2e8d1f9b05",
      "source": "cloned",
      "providerVoiceId": "vc_7f2a9c41e8b3",
      "name": "Priya narrator",
      "description": "Warm, even delivery for lessons",
      "gender": "female",
      "prompt": null,
      "createdAt": "2026-09-15T09:12:40.118204+00:00",
      "consentVerifiedAt": "2026-09-15T09:14:02.551930+00:00"
    }
  ],
  "limit": 10,
  "creditsPerVoice": 10000,
  "minSeconds": 20,
  "maxSeconds": 300
}

limit, creditsPerVoice, minSeconds and maxSeconds describe voice cloning: up to 10 cloned voices per workspace, 10,000 credits each, recordings of 20 to 300 seconds.

Credits

Free.

Example

bash

curl "https://api.cinara.ai/v1/voices?source=cloned" \
  -H "Authorization: Bearer $CINARA_API_KEY"

Update a voice

PATCH /v1/voices/{id}

Renames or re-describes one of the workspace's voices. A cloned voice's consent record never changes.

Request body

Send at least one field.

Field	Type	Description
`name`	string	1–40 characters.
`description`	string	Up to 120 characters. An empty string clears it.
`gender`	string or null	`female`, `male`, or `null` to clear it.

Response

200 OK

json

{
  "voice": {
    "id": "5a0e3c7b-91d2-4f68-b3a4-6c2e8d1f9b05",
    "source": "cloned",
    "providerVoiceId": "vc_7f2a9c41e8b3",
    "name": "Priya, lessons",
    "description": "Warm, even delivery for lessons",
    "gender": "female",
    "prompt": null,
    "createdAt": "2026-09-15T09:12:40.118204+00:00",
    "consentVerifiedAt": "2026-09-15T09:14:02.551930+00:00"
  }
}

Errors

Status	`error`	Message
404	`not_found`	`Voice not found`
400	`invalid_request`	`Give the voice a name of up to 40 characters`
400	`invalid_request`	`Keep the description under 120 characters`
400	`invalid_request`	`The voice type must be female or male`
400	`invalid_request`	`Nothing to change`
400	`invalid_request`	`Cinara doesn't name voices after well-known people.`

Credits

Free.

Example

bash

curl -X PATCH "https://api.cinara.ai/v1/voices/5a0e3c7b-91d2-4f68-b3a4-6c2e8d1f9b05" \
  -H "Authorization: Bearer $CINARA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"name": "Priya, lessons"}'

Delete a voice

DELETE /v1/voices/{id}

Deletes one of the workspace's voices, its favourites, and its preview and sample audio. For a cloned voice, the cloning job keeps its consent details.

Response

200 OK

json

{ "ok": true }

Errors

Status	`error`	Message
404	`not_found`	`Voice not found`

Credits

Free.

Example

bash

curl -X DELETE "https://api.cinara.ai/v1/voices/5a0e3c7b-91d2-4f68-b3a4-6c2e8d1f9b05" \
  -H "Authorization: Bearer $CINARA_API_KEY"

Get a workspace voice's preview

GET /v1/voices/{id}/preview

Returns the preview clip of one of the workspace's voices, by its UUID.

Response

200 OK with Content-Type: audio/mpeg and Cache-Control: private, max-age=3600.

For a cloned voice the clip says "Hi, this is name. This is how my voice sounds in Cinara." For a designed voice it's the design's preview.

Errors

Status	`error`	Message
404	`not_found`	`Preview not found`

Credits

Free.

Example

bash

curl "https://api.cinara.ai/v1/voices/5a0e3c7b-91d2-4f68-b3a4-6c2e8d1f9b05/preview" \
  -H "Authorization: Bearer $CINARA_API_KEY" -o preview.mp3

List favourite voices

GET /v1/voices/favourites

Lists the voice ids the key's creator marked as favourites in this workspace, newest first. Favourites are per member, per workspace.

Response

200 OK

json

{ "voiceIds": ["mentor", "vc_7f2a9c41e8b3", "haven"] }

Ids are library voice ids or workspace voices' providerVoiceIds.

Credits

Free.

Example

bash

curl https://api.cinara.ai/v1/voices/favourites \
  -H "Authorization: Bearer $CINARA_API_KEY"

Add a favourite voice

PUT /v1/voices/favourites/{voiceId}

Marks a voice as a favourite. Adding one that's already a favourite succeeds.

Path parameters

Name	Description
`voiceId`	A library voice id, or a workspace voice's `providerVoiceId`.

Response

200 OK

json

{ "ok": true }

Errors

Status	`error`	Message
404	`not_found`	`Voice not found`
400	`invalid_request`	`You can keep up to 200 favourite voices`

Credits

Free.

Example

bash

curl -X PUT https://api.cinara.ai/v1/voices/favourites/mentor \
  -H "Authorization: Bearer $CINARA_API_KEY"

Remove a favourite voice

DELETE /v1/voices/favourites/{voiceId}

Removes a voice from favourites. It succeeds even when the voice wasn't a favourite.

Response

200 OK

json

{ "ok": true }

Credits

Free.

Example

bash

curl -X DELETE https://api.cinara.ai/v1/voices/favourites/mentor \
  -H "Authorization: Bearer $CINARA_API_KEY"

Voice design

Voice design turns a description into a new voice with a spoken preview. Each design is a background job. Listen to the preview, then save the designs you like to the workspace's voices. A design must be saved within 7 days.

Get voice design options

GET /v1/voice-design/options

Response

200 OK

json

{
  "creditsPerDesign": 20000,
  "promptMin": 20,
  "promptMax": 2000,
  "previewMin": 50,
  "previewMax": 500,
  "keepDays": 7,
  "maxVoices": 30,
  "samples": [
    {
      "id": "conversation",
      "label": "Conversation",
      "text": "Hey, it's so good to hear from you! I was just thinking about that trip we planned. Are we still going next weekend, or should we wait until the weather gets a little warmer? Either way, I'll bring the snacks."
    }
  ],
  "traits": {
    "gender": ["female", "male"],
    "age": ["child", "teenager", "young adult", "middle-aged", "elderly"],
    "accent": ["American", "British", "Australian", "Canadian", "Irish", "Scottish", "Indian", "Nigerian", "South African", "French", "Spanish", "German"],
    "pace": ["slow", "measured", "fast"],
    "tone": ["warm", "calm", "deep", "bright", "energetic", "gentle", "authoritative", "playful", "raspy", "breathy", "mysterious", "cheerful"]
  }
}

samples holds ready-made preview texts with the ids story, conversation, advert, character, news, meditation and hindi (shortened above). traits are words that help when writing a description.

Credits

Free.

Example

bash

curl https://api.cinara.ai/v1/voice-design/options \
  -H "Authorization: Bearer $CINARA_API_KEY"

Design a voice

POST /v1/voice-designs

Starts designing a voice from a description.

Needs a verified phone.

Request body

Field	Type	Required	Description
`prompt`	string	yes	A description of the voice. 20–2,000 characters. Describe the voice itself; naming well-known people is refused.
`previewText`	string	yes	What the new voice says in its preview. 50–500 characters.

Response

201 Created

json

{
  "generation": {
    "id": "8e3d1b7a-4c2f-4a6e-9b0d-7f5c2a1e3d98",
    "module": "voice_design",
    "status": "running",
    "input": {
      "prompt": "An elderly man with a warm, deep voice and a British accent, speaking slowly.",
      "previewText": "Come closer and let me tell you a story. Long ago, in a small village by the sea, there lived a fisherman who could hear the waves sing."
    },
    "output": { "stage": "queued", "step": 1, "stageLabel": "Designing the voice" },
    "credits": 20000,
    "error": null,
    "created_at": "2026-09-15T13:20:11.402113+00:00",
    "completed_at": null
  },
  "credits": 39756
}

Errors

Status	`error`	Message
400	`invalid_request`	`Describe the voice in at least 20 characters`
400	`invalid_request`	`Keep the description under 2,000 characters`
400	`invalid_request`	`The preview text needs at least 50 characters`
400	`invalid_request`	`The preview text can be up to 500 characters`
400	`invalid_request`	`Cinara doesn't design voices that copy well-known people. Describe the voice itself instead.`
402	`insufficient_credits`	`This needs 20,000 credits`
403	`phone_unverified`	`Verify your phone number to start creating.`
502	`generation_failed`	`Voice design couldn't start. Your credits were refunded.`

Credits

20,000 credits per design, whether or not you save it. Refunded if the design fails. Saving is free.

Example

bash

curl -X POST https://api.cinara.ai/v1/voice-designs \
  -H "Authorization: Bearer $CINARA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "An elderly man with a warm, deep voice and a British accent, speaking slowly.", "previewText": "Come closer and let me tell you a story. Long ago, in a small village by the sea, there lived a fisherman who could hear the waves sing."}'

Get a voice design

GET /v1/voice-designs/{id}

Checks a design, moves it forward, and returns it. Poll until status isn't running.

Response

200 OK

json

{
  "generation": {
    "id": "8e3d1b7a-4c2f-4a6e-9b0d-7f5c2a1e3d98",
    "module": "voice_design",
    "status": "succeeded",
    "input": {
      "prompt": "An elderly man with a warm, deep voice and a British accent, speaking slowly.",
      "previewText": "Come closer and let me tell you a story. Long ago, in a small village by the sea, there lived a fisherman who could hear the waves sing."
    },
    "output": {
      "design": {
        "providerVoiceId": "vd_19c4e2a7b6f0"
      }
    },
    "credits": 20000,
    "error": null,
    "created_at": "2026-09-15T13:20:11.402113+00:00",
    "completed_at": "2026-09-15T13:20:39.771640+00:00"
  },
  "voice": null,
  "expiresAt": "2026-09-22T13:20:11.402Z",
  "credits": 39756
}

Field	Description
`voice`	The saved voice object, or `null` until the design is saved.
`expiresAt`	The last moment the design can be saved (7 days after it was created).
`credits`	The new balance. Present only on the read where the status changed.

Once saved, output.savedVoiceId holds the voice's UUID.

A failed design has status: "failed" and an error such as Voice design took too long and was stopped. or Voice design failed.

Errors

Status	`error`	Message
404	`not_found`	`Not found`

Credits

Free.

Polling example

bash

while :; do
  RES=$(curl -s "https://api.cinara.ai/v1/voice-designs/$ID" -H "Authorization: Bearer $CINARA_API_KEY")
  [ "$(echo "$RES" | jq -r .generation.status)" != "running" ] && break
  sleep 5
done
curl -s "https://api.cinara.ai/v1/voice-designs/$ID/preview" -H "Authorization: Bearer $CINARA_API_KEY" -o design.mp3

Get a voice design's preview

GET /v1/voice-designs/{id}/preview

Returns the design's spoken preview.

Response

200 OK with the clip's content type (normally audio/mpeg) and Cache-Control: private, max-age=3600.

Errors

Status	`error`	Message
404	`not_found`	`Preview not found`

Credits

Free.

Example

bash

curl "https://api.cinara.ai/v1/voice-designs/8e3d1b7a-4c2f-4a6e-9b0d-7f5c2a1e3d98/preview" \
  -H "Authorization: Bearer $CINARA_API_KEY" -o design.mp3

Save a voice design

POST /v1/voice-designs/{id}/save

Saves a finished design to the workspace's voices, where speech, dialogue, voice changer and the other features can use it.

Needs a verified phone.

Saving a design that's already saved returns the existing voice with 200.

Request body

Field	Type	Required	Description
`name`	string	yes	1–40 characters.
`description`	string	no	Up to 120 characters; longer is refused. Defaults to the first 120 characters of the design's prompt.
`gender`	string	no	`female` or `male`.

A workspace can have up to 30 designed voices.

Response

201 Created (or 200 OK when already saved)

json

{
  "voice": {
    "id": "d71f0c3e-8a2b-4e5d-b6c9-0a4e1f7b2c85",
    "source": "designed",
    "providerVoiceId": "vd_19c4e2a7b6f0",
    "name": "Old storyteller",
    "description": "An elderly man with a warm, deep voice and a British accent, speaking slowly.",
    "gender": "male",
    "prompt": "An elderly man with a warm, deep voice and a British accent, speaking slowly.",
    "createdAt": "2026-09-15T13:25:02.019873+00:00",
    "consentVerifiedAt": null
  }
}

Errors

Status	`error`	Message
404	`not_found`	`Not found`
400	`invalid_request`	`This design isn't ready yet.`
410	`expired`	`Designs must be saved within 7 days. Generate this one again to save it.`
400	`invalid_request`	`Give the voice a name of up to 40 characters`
400	`invalid_request`	`Keep the description under 120 characters`
400	`invalid_request`	`Cinara doesn't design voices that copy well-known people.`
400	`invalid_request`	`A workspace can have up to 30 designed voices. Delete one to save another.`
404	`not_found`	`The preview is missing. Generate the voice again.`
403	`phone_unverified`	`Verify your phone number to start creating.`
502	`generation_failed`	`Couldn't save the voice. Try again in a moment.`

Credits

Free. The design was already paid for.

Example

bash

curl -X POST "https://api.cinara.ai/v1/voice-designs/8e3d1b7a-4c2f-4a6e-9b0d-7f5c2a1e3d98/save" \
  -H "Authorization: Bearer $CINARA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"name": "Old storyteller", "gender": "male"}'

Voice cloning

A voice is cloned only from a recording that starts with a one-time consent statement, read aloud by the person whose voice it is. This stops cloning someone from recordings that already exist.

The flow:

Get a statement. POST /v1/voices/consent with the speaker's full name. You get a challengeId and a phrase. It expires after 30 minutes and works for one recording.
Record. The speaker reads the phrase word for word, including the code words, then keeps talking naturally. Only that one person should be heard. The recording needs at least 20 seconds and can be up to 5 minutes.
Upload the recording with /v1/uploads. It must be an audio file.
Clone. POST /v1/voices with the upload's key, the challengeId, a name and attested: true.
Poll GET /v1/voice-clones/{id} until it finishes. The response then includes the new voice.

The consent check passes when:

every code word is heard, in order;
at least 80% of the statement's words are heard, in order (small misspellings are allowed);
only one speaker is heard (a speaker counts when they say at least 15% of the words);
the speech heard lasts at least 20 seconds.

Names of well-known people are refused in the speaker name, voice name and description.

POST /v1/voices/consent

Creates a one-time consent statement for a speaker to read.

Needs a verified phone.

Request body

Field	Type	Required	Description
`speakerName`	string	yes	The speaker's full name. At least 2 characters. Longer than 60 characters is cut to 60.

Response

201 Created

json

{
  "challengeId": "b3c8e1f4-7a2d-4e9b-8c5a-2f6d0e9b1a37",
  "phrase": "My name is Priya Raman. I agree to let Cinara create a digital copy of my voice. My one-time code is lantern cobalt meadow violin.",
  "expiresAt": "2026-09-15T09:40:00.000Z"
}

The phrase always has this form: My name is <speakerName>. I agree to let Cinara create a digital copy of my voice. My one-time code is <four code words>.

Errors

Status	`error`	Message
400	`invalid_request`	`Enter the speaker's full name`
400	`invalid_request`	`Cinara doesn't clone the voices of well-known people.`
403	`phone_unverified`	`Verify your phone number to start creating.`

Credits

Free.

Example

bash

curl -X POST https://api.cinara.ai/v1/voices/consent \
  -H "Authorization: Bearer $CINARA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"speakerName": "Priya Raman"}'

Clone a voice

POST /v1/voices

Starts cloning a voice from an uploaded recording that begins with the consent statement.

Needs a verified phone.

Request body

Field	Type	Required	Description
`attested`	boolean	yes	Must be `true`: you are this speaker, or have their permission.
`key`	string	yes	The storage key of a finished upload in this workspace. Must be an audio file.
`challengeId`	string	yes	From Get a consent statement. Must be unused, unexpired, and created by the same member (the key's creator) in this workspace.
`name`	string	yes	The voice's name. Longer than 40 characters is cut to 40.
`durationSeconds`	number	yes	The recording's length in seconds. 20–300.
`description`	string	no	Longer than 120 characters is cut to 120.
`gender`	string	no	`female` or `male`.
`removeNoise`	boolean	no	Reduce background noise before cloning. Default `false`.
`fileName`	string	no	A display name for the recording. Up to 200 characters. Defaults to the upload's file name.

A workspace can have up to 10 cloned voices.

Response

201 Created

json

{
  "generation": {
    "id": "1f9c5e2a-3b7d-4a8c-9e6f-0d2b4c7a1e59",
    "module": "voice_clone",
    "status": "running",
    "input": {
      "key": "workspaces/3f6b2a1c-8d4e-4b7a-9c2d-5e1f0a9b8c7d/uploads/4c1a7e3b-9d2f-4b8e-a6c0-5e3f1d9b7a24/priya-consent.m4a",
      "fileName": "priya-consent.m4a",
      "contentType": "audio/mp4",
      "size": 1843200,
      "durationSeconds": 74.2,
      "name": "Priya narrator",
      "description": "Warm, even delivery for lessons",
      "gender": "female",
      "speakerName": "Priya Raman",
      "challengeId": "b3c8e1f4-7a2d-4e9b-8c5a-2f6d0e9b1a37",
      "removeNoise": false
    },
    "output": { "stage": "queued", "step": 1, "stageLabel": "Checking the consent statement" },
    "credits": 10000,
    "error": null,
    "created_at": "2026-09-15T09:13:21.660419+00:00",
    "completed_at": null
  },
  "credits": 44756
}

Errors

Status	`error`	Message
400	`invalid_request`	`Confirm that you are this speaker, or have their permission.`
400	`invalid_request`	`Upload the recording first`
400	`invalid_request`	`Give the voice a name`
400	`invalid_request`	`Cinara doesn't clone the voices of well-known people.`
400	`invalid_request`	`The recording needs at least 20 seconds: the statement, then natural speech.`
400	`invalid_request`	`Recordings can be up to 5 minutes`
400	`invalid_request`	`This consent statement has expired or was already used. Get a new one.`
400	`invalid_request`	`A workspace can have up to 10 cloned voices. Delete one to make another.`
404	`not_found`	`That upload isn't finished. Upload the recording again.`
400	`invalid_request`	`Upload an audio recording`
400	`invalid_request`	`This consent statement was already used. Get a new one.`
402	`insufficient_credits`	`This needs 10,000 credits`
403	`phone_unverified`	`Verify your phone number to start creating.`
502	`generation_failed`	`Voice cloning couldn't start. Your credits were refunded.`

If the request fails with 402 or 502, the consent statement is released and can be used again.

Credits

10,000 credits per cloned voice. Refunded if the consent check or cloning fails. Speaking with the voice afterwards costs normal speech credits.

Example

bash

curl -X POST https://api.cinara.ai/v1/voices \
  -H "Authorization: Bearer $CINARA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "attested": true,
    "key": "workspaces/3f6b2a1c-8d4e-4b7a-9c2d-5e1f0a9b8c7d/uploads/4c1a7e3b-9d2f-4b8e-a6c0-5e3f1d9b7a24/priya-consent.m4a",
    "challengeId": "b3c8e1f4-7a2d-4e9b-8c5a-2f6d0e9b1a37",
    "name": "Priya narrator",
    "gender": "female",
    "durationSeconds": 74.2
  }'

Get a voice cloning job

GET /v1/voice-clones/{id}

Checks a cloning job, moves it forward, and returns it. Poll until status isn't running.

Steps: checking the consent statement, preparing the recording (for formats other than MP3, WAV and M4A), creating the voice, saving the voice.

Response

200 OK

json

{
  "generation": {
    "id": "1f9c5e2a-3b7d-4a8c-9e6f-0d2b4c7a1e59",
    "module": "voice_clone",
    "status": "succeeded",
    "input": {
      "key": "workspaces/3f6b2a1c-8d4e-4b7a-9c2d-5e1f0a9b8c7d/uploads/4c1a7e3b-9d2f-4b8e-a6c0-5e3f1d9b7a24/priya-consent.m4a",
      "fileName": "priya-consent.m4a",
      "contentType": "audio/mp4",
      "size": 1843200,
      "durationSeconds": 74.2,
      "name": "Priya narrator",
      "description": "Warm, even delivery for lessons",
      "gender": "female",
      "speakerName": "Priya Raman",
      "challengeId": "b3c8e1f4-7a2d-4e9b-8c5a-2f6d0e9b1a37",
      "removeNoise": false
    },
    "output": {
      "voiceId": "5a0e3c7b-91d2-4f68-b3a4-6c2e8d1f9b05",
      "consent": { "matchedShare": 0.958, "codeHeard": true, "speakers": 1, "heardSeconds": 72.9 },
      "step": 2
    },
    "credits": 10000,
    "error": null,
    "created_at": "2026-09-15T09:13:21.660419+00:00",
    "completed_at": "2026-09-15T09:14:02.551930+00:00"
  },
  "voice": {
    "id": "5a0e3c7b-91d2-4f68-b3a4-6c2e8d1f9b05",
    "source": "cloned",
    "providerVoiceId": "vc_7f2a9c41e8b3",
    "name": "Priya narrator",
    "description": "Warm, even delivery for lessons",
    "gender": "female",
    "prompt": null,
    "createdAt": "2026-09-15T09:14:02.201187+00:00",
    "consentVerifiedAt": "2026-09-15T09:14:02.551930+00:00"
  },
  "credits": 44756
}

output.voiceId is the new voice's UUID, not the id to speak with. Use voice.providerVoiceId to speak.
output.consent shows the check: share of the statement heard, whether the code was heard, speakers heard, and seconds of speech heard.
voice is null until the voice is saved.

If the consent check fails, status is failed, output.consent shows the check and error is one of:

The one-time code wasn't heard. Read the whole statement aloud, including the code words, then keep talking.
The consent statement wasn't read clearly enough. Read it word for word, then keep talking.
More than one voice was heard. Record only the person whose voice is being cloned.
Keep talking for at least 20 seconds in total, so there's enough of the voice to clone.
The consent statement is missing. Start again.

Other failures: Voice cloning took too long and was stopped., This file couldn't be read. Try MP3, WAV, M4A, MP4 or WebM., Voice cloning failed.

Errors

Status	`error`	Message
404	`not_found`	`Not found`

Credits

Free to read.

Polling example

bash

while :; do
  RES=$(curl -s "https://api.cinara.ai/v1/voice-clones/$ID" -H "Authorization: Bearer $CINARA_API_KEY")
  STATUS=$(echo "$RES" | jq -r .generation.status)
  [ "$STATUS" != "running" ] && break
  echo "$(echo "$RES" | jq -r .generation.output.stageLabel)…"
  sleep 5
done
echo "$RES" | jq '{status: .generation.status, error: .generation.error, speakWith: .voice.providerVoiceId}'

Voices ​

Two ids per voice ​

The voice object ​

List workspace voices ​

Query parameters ​

Response ​

Credits ​

Example ​

Update a voice ​

Request body ​

Response ​

Errors ​

Credits ​

Example ​

Delete a voice ​

Response ​

Errors ​

Credits ​

Example ​

Get a workspace voice's preview ​

Response ​

Errors ​

Credits ​

Example ​

List favourite voices ​

Response ​

Credits ​

Example ​

Add a favourite voice ​

Path parameters ​

Response ​

Errors ​

Credits ​

Example ​

Remove a favourite voice ​

Response ​

Credits ​

Example ​

Voice design ​

Get voice design options ​

Response ​

Credits ​

Example ​

Design a voice ​

Request body ​

Response ​

Errors ​

Credits ​

Example ​

Get a voice design ​

Response ​

Errors ​

Credits ​

Polling example ​

Get a voice design's preview ​

Response ​

Errors ​

Credits ​

Example ​

Save a voice design ​

Request body ​

Response ​

Errors ​

Credits ​

Example ​

Voice cloning ​

Get a consent statement ​

Request body ​

Response ​

Errors ​

Credits ​

Example ​

Clone a voice ​

Request body ​

Response ​

Errors ​

Credits ​

Example ​

Get a voice cloning job ​

Response ​

Voices

Two ids per voice

The voice object

List workspace voices

Query parameters

Response

Credits

Example

Update a voice

Request body

Response

Errors

Credits

Example

Delete a voice

Response

Errors

Credits

Example

Get a workspace voice's preview

Response

Errors

Credits

Example

List favourite voices

Response

Credits

Example

Add a favourite voice

Path parameters

Response

Errors

Credits

Example

Remove a favourite voice

Response

Credits

Example

Voice design

Get voice design options

Response

Credits

Example

Design a voice

Request body

Response

Errors

Credits

Example

Get a voice design

Response

Errors

Credits

Polling example

Get a voice design's preview

Response

Errors

Credits

Example

Save a voice design

Request body

Response

Errors

Credits

Example

Voice cloning

Get a consent statement

Request body

Response

Errors

Credits

Example

Clone a voice

Request body

Response

Errors

Credits

Example

Get a voice cloning job

Response