Voice Cloning

Create custom voices from audio samples — instantly from a single recording, or as a deep clone fine-tuned on a training dataset — then share them across your company and use them as any agent's voice.

All paths on this page are relative to https://api.autessa.com/autessa_poweruser/voice. Every request is a POST with a JSON body, and every ID is an opaque string. See the Poweruser API overview for authentication and conventions.

Instant clones vs deep clones

There are two ways to clone a voice, with different quality/effort trade-offs and separate company entitlements:

Instant clones are created synchronously from a single WAV sample in one /instant-clone/create call. Available when voice cloning is enabled for the company.
Deep clones are fine-tuned voice models trained on a dataset of samples. They follow a three-step lifecycle — /deep-clone/create-dataset, then /deep-clone/upload-files, then /deep-clone/start-finetune — and train asynchronously. Poll /deep-clone/get-status until the clone reaches COMPLETED. Available only when voice fine-tuning is enabled for the company (a separate, higher-tier entitlement).

If the relevant entitlement is missing, the create call returns a 400 with "Voice cloning is not enabled for this company" (instant) or "Voice fine-tuning is not enabled for this company" (deep).

A deep clone moves through these statuses: PENDING (dataset created) → UPLOADING (samples being added) → TRAINING (fine-tune running) → COMPLETED or FAILED.

Once an instant clone is created or a deep clone completes, the voice appears under customVoiceNames in the voice options returned by the Agents API and can be set as an agent's voiceConfig.voiceName.

Both clone types use the platform-wide sharing model. Endpoints enforce the linked user's access level on the clone:

Operation	Required access
`get`, `get-all`, `get-status`, `get-sharing-details`	Any access (READ or higher)
`upload-files`, `start-finetune`	WRITE or ADMIN
`share`, `delete`, `transfer-ownership`	ADMIN

Generate an audio upload link

POST/voice/generate-audio-upload-link

Returns a pre-signed S3 URL for uploading a WAV voice sample. Upload the file with an HTTP PUT to uploadUrl (with Content-Type: audio/wav), then pass the returned s3Uri as the sample reference to /instant-clone/create or /deep-clone/upload-files.

Request one link per file — each call returns a unique upload destination.

Request parameters

None — send an empty object {} (WAV format is currently fixed).

Response

Name
uploadUrl
Type
string
Description
Pre-signed URL; PUT the file bytes here with Content-Type: audio/wav.
Name
s3Uri
Type
string
Description
S3 URI referencing the uploaded file. Pass this to the clone endpoints.
Name
expiresInSeconds
Type
integer
Description
How long the upload URL stays valid (currently 600 — ten minutes).
Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
JavaScript
Python

curl https://api.autessa.com/autessa_poweruser/voice/generate-audio-upload-link \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{}'

# Then upload the sample:
curl -X PUT "$UPLOAD_URL" \
  -H "Content-Type: audio/wav" \
  --data-binary '@voice-sample.wav'

const response = await fetch(
  'https://api.autessa.com/autessa_poweruser/voice/generate-audio-upload-link',
  {
    method: 'POST',
    headers: {
      'Authorization': process.env.POWERUSER_KEY,
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({})
  }
);

const { uploadUrl, s3Uri } = await response.json();

// Upload the WAV bytes
await fetch(uploadUrl, {
  method: 'PUT',
  headers: { 'Content-Type': 'audio/wav' },
  body: wavFileBytes
});

import os
import requests

response = requests.post(
    'https://api.autessa.com/autessa_poweruser/voice/generate-audio-upload-link',
    headers={'Authorization': os.environ['POWERUSER_KEY']},
    json={}
)
link = response.json()

with open('voice-sample.wav', 'rb') as f:
    requests.put(
        link['uploadUrl'],
        headers={'Content-Type': 'audio/wav'},
        data=f
    )

s3_uri = link['s3Uri']

Response
{
  "uploadUrl": "https://s3.amazonaws.com/autessa-voice/...",
  "s3Uri": "s3://autessa-voice/voice-clones/audio/prod/1234/8f3a2b.wav",
  "expiresInSeconds": 600,
  "errors": []
}

Create an instant clone

POST/voice/instant-clone/create

Creates a voice clone from a single uploaded WAV sample. The clone is created synchronously — the response already carries the provider voice ID and the voice is immediately usable. New clones start PRIVATE; use /instant-clone/share to grant access.

Requires the company's voice cloning entitlement; without it the call returns a 400 with "Voice cloning is not enabled for this company".

Request parameters

Name
voiceName
Type
string
Description
Display name for the voice. Must be non-blank.
Name
audioS3Uri
Type
string
Description
S3 URI of the uploaded WAV sample, from /generate-audio-upload-link.

Response

Name
voiceCloneId
Type
string
Description
The new instant clone's ID.
Name
voiceName
Type
string
Description
The stored display name.
Name
serviceVoiceId
Type
string
Description
The underlying speech provider's voice identifier.
Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
JavaScript
Python

curl https://api.autessa.com/autessa_poweruser/voice/instant-clone/create \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "voiceName": "Support Narrator",
    "audioS3Uri": "s3://autessa-voice/voice-clones/audio/prod/1234/8f3a2b.wav"
  }'

const response = await fetch(
  'https://api.autessa.com/autessa_poweruser/voice/instant-clone/create',
  {
    method: 'POST',
    headers: {
      'Authorization': process.env.POWERUSER_KEY,
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      voiceName: 'Support Narrator',
      audioS3Uri: 's3://autessa-voice/voice-clones/audio/prod/1234/8f3a2b.wav'
    })
  }
);

const { voiceCloneId } = await response.json();

import os
import requests

response = requests.post(
    'https://api.autessa.com/autessa_poweruser/voice/instant-clone/create',
    headers={'Authorization': os.environ['POWERUSER_KEY']},
    json={
        'voiceName': 'Support Narrator',
        'audioS3Uri': 's3://autessa-voice/voice-clones/audio/prod/1234/8f3a2b.wav'
    }
)

voice_clone_id = response.json()['voiceCloneId']

Response
{
  "voiceCloneId": "V1StGXR8Z5",
  "voiceName": "Support Narrator",
  "serviceVoiceId": "a0e99841-438c-4a64-b679-ae501e7d6091",
  "errors": []
}

Get an instant clone

POST/voice/instant-clone/get

Retrieves a single instant clone, including its sharing state and the calling user's access level. Requires any access to the clone.

Request parameters

Name
instantVoiceCloneId
Type
string
Description
The instant clone to fetch.

Response

Name
voiceCloneId
Type
string
Description
The clone's ID.
Name
voiceName
Type
string
Description
Display name.
Name
serviceVoiceId
Type
string
Description
The speech provider's voice identifier.
Name
sharingLevel
Type
string
Description
PRIVATE, SPECIFIC_MEMBER, or COMPANY.
Name
companyPermissionLevel
Type
string
Description
Access level granted company-wide, when sharingLevel is COMPANY: READ, WRITE, PUBLISH, or ADMIN.
Name
userAccessLevel
Type
string
Description
The calling user's effective access level on this clone.
Name
createdAt
Type
string
Description
ISO-8601 creation timestamp.
Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/instant-clone/get \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "instantVoiceCloneId": "V1StGXR8Z5" }'

Response
{
  "voiceCloneId": "V1StGXR8Z5",
  "voiceName": "Support Narrator",
  "serviceVoiceId": "a0e99841-438c-4a64-b679-ae501e7d6091",
  "sharingLevel": "PRIVATE",
  "companyPermissionLevel": null,
  "userAccessLevel": "ADMIN",
  "createdAt": "2026-06-28T17:42:10",
  "errors": []
}

List all instant clones

POST/voice/instant-clone/get-all

Lists every instant clone visible to the linked user, grouped by how they have access.

Request parameters

None — send an empty object {}.

Response

Each entry is a summary with id, voiceName, serviceVoiceId, sharingLevel, and createdAt.

Name
ownedClones
Type
array<VoiceCloneSummary>
Description
Instant clones the linked user owns.
Name
companySharedClones
Type
array<VoiceCloneSummary>
Description
Clones shared with the whole company by other users.
Name
sharedWithMeClones
Type
array<VoiceCloneSummary>
Description
Clones shared specifically with the linked user.
Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/instant-clone/get-all \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{}'

Response
{
  "ownedClones": [
    {
      "id": "V1StGXR8Z5",
      "voiceName": "Support Narrator",
      "serviceVoiceId": "a0e99841-438c-4a64-b679-ae501e7d6091",
      "sharingLevel": "PRIVATE",
      "createdAt": "2026-06-28T17:42:10"
    }
  ],
  "companySharedClones": [],
  "sharedWithMeClones": [],
  "errors": []
}

Delete an instant clone

POST/voice/instant-clone/delete

Permanently deletes an instant clone. Agents configured to use the voice fall back to their built-in voice.

Requires ADMIN access on the clone.

Request parameters

Name
instantVoiceCloneId
Type
string
Description
The instant clone to delete.

Response

Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/instant-clone/delete \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "instantVoiceCloneId": "V1StGXR8Z5" }'

Response
{
  "errors": []
}

POST/voice/instant-clone/share

Sets the clone's sharing level and grants per-user access. When shareLevel is SPECIFIC_MEMBER, userIds and permissionLevels are parallel arrays — the user at index i receives the permission at index i (defaulting to READ when permissionLevels is shorter or omitted).

Requires ADMIN access on the clone.

Request parameters

Name
instantVoiceCloneId
Type
string
Description
The instant clone to share.
Name
shareLevel
Type
string
Description
PRIVATE, SPECIFIC_MEMBER, or COMPANY.
Name
companyPermissionLevel
Type
string
Description
When shareLevel is COMPANY, the access level every company member gets: READ, WRITE, PUBLISH, or ADMIN.
Name
userIds
Type
array<string>
Description
When shareLevel is SPECIFIC_MEMBER, the users to grant access to.
Name
permissionLevels
Type
array<string>
Description
Access level for each entry in userIds, positionally matched. Missing entries default to READ.

Response

Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/instant-clone/share \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "instantVoiceCloneId": "V1StGXR8Z5",
    "shareLevel": "SPECIFIC_MEMBER",
    "userIds": ["uB7nT2wQ9k"],
    "permissionLevels": ["READ"]
  }'

Response
{
  "errors": []
}

POST/voice/instant-clone/get-sharing-details

Returns the clone's current sharing configuration and the calling user's own access level. Requires any access to the clone.

Request parameters

Name
instantVoiceCloneId
Type
string
Description
The instant clone to inspect.

Response

Name
sharingLevel
Type
string
Description
PRIVATE, SPECIFIC_MEMBER, or COMPANY.
Name
companyPermissionLevel
Type
string
Description
Access level granted company-wide, when sharingLevel is COMPANY.
Name
sharedWith
Type
array<UserShare>
Description
Per-user grants: userEmail, userId (string), accessLevel.
Name
userAccessLevel
Type
string
Description
The calling user's effective access level.
Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/instant-clone/get-sharing-details \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "instantVoiceCloneId": "V1StGXR8Z5" }'

Response
{
  "sharingLevel": "SPECIFIC_MEMBER",
  "companyPermissionLevel": null,
  "sharedWith": [
    {
      "userEmail": "sam@acme.com",
      "userId": "uB7nT2wQ9k",
      "accessLevel": "READ"
    }
  ],
  "userAccessLevel": "ADMIN",
  "errors": []
}

Transfer instant clone ownership

POST/voice/instant-clone/transfer-ownership

Transfers the clone to another member of the company. Requires ADMIN access on the clone.

Request parameters

Name
instantVoiceCloneId
Type
string
Description
The instant clone to transfer.
Name
newOwnerId
Type
string
Description
User ID of the new owner.

Response

Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/instant-clone/transfer-ownership \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "instantVoiceCloneId": "V1StGXR8Z5",
    "newOwnerId": "uB7nT2wQ9k"
  }'

Response
{
  "errors": []
}

Create a deep clone dataset

POST/voice/deep-clone/create-dataset

Starts a deep clone by creating its training dataset. The new clone starts in status PENDING with sharing level PRIVATE. Next, add samples with /deep-clone/upload-files.

Requires the company's voice fine-tuning entitlement; without it the call returns a 400 with "Voice fine-tuning is not enabled for this company".

Request parameters

Name
voiceName
Type
string
Description
Display name for the voice. Must be non-blank.

Response

Name
voiceCloneId
Type
string
Description
The new deep clone's ID. Use it as deepVoiceCloneId in the rest of the deep-clone endpoints.
Name
datasetId
Type
string
Description
The speech provider's dataset identifier.
Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
JavaScript
Python

curl https://api.autessa.com/autessa_poweruser/voice/deep-clone/create-dataset \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "voiceName": "Brand Voice" }'

const response = await fetch(
  'https://api.autessa.com/autessa_poweruser/voice/deep-clone/create-dataset',
  {
    method: 'POST',
    headers: {
      'Authorization': process.env.POWERUSER_KEY,
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({ voiceName: 'Brand Voice' })
  }
);

const { voiceCloneId } = await response.json();

import os
import requests

response = requests.post(
    'https://api.autessa.com/autessa_poweruser/voice/deep-clone/create-dataset',
    headers={'Authorization': os.environ['POWERUSER_KEY']},
    json={'voiceName': 'Brand Voice'}
)

voice_clone_id = response.json()['voiceCloneId']

Response
{
  "voiceCloneId": "dV7kM3pQ9w",
  "datasetId": "ds_4f8c1a2b9d",
  "errors": []
}

Upload files to a dataset

POST/voice/deep-clone/upload-files

Adds WAV training samples to the clone's dataset and moves the clone to status UPLOADING. Upload each sample first via /generate-audio-upload-link and pass the resulting S3 URIs. If any upload fails, the clone is marked FAILED and the message is returned in errors.

Requires WRITE or ADMIN access on the clone, and that the dataset exists ("Dataset not created yet" otherwise).

Request parameters

Name
deepVoiceCloneId
Type
string
Description
The deep clone whose dataset to fill.
Name
audioS3Uris
Type
array<string>
Description
S3 URIs of the uploaded WAV samples.

Response

Name
filesUploaded
Type
integer
Description
Number of files added to the dataset in this call.
Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
JavaScript
Python

curl https://api.autessa.com/autessa_poweruser/voice/deep-clone/upload-files \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "deepVoiceCloneId": "dV7kM3pQ9w",
    "audioS3Uris": [
      "s3://autessa-voice/voice-clones/audio/prod/1234/8f3a2b.wav",
      "s3://autessa-voice/voice-clones/audio/prod/1234/2c9d7e.wav"
    ]
  }'

const response = await fetch(
  'https://api.autessa.com/autessa_poweruser/voice/deep-clone/upload-files',
  {
    method: 'POST',
    headers: {
      'Authorization': process.env.POWERUSER_KEY,
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      deepVoiceCloneId: 'dV7kM3pQ9w',
      audioS3Uris: [
        's3://autessa-voice/voice-clones/audio/prod/1234/8f3a2b.wav',
        's3://autessa-voice/voice-clones/audio/prod/1234/2c9d7e.wav'
      ]
    })
  }
);

const { filesUploaded } = await response.json();

import os
import requests

response = requests.post(
    'https://api.autessa.com/autessa_poweruser/voice/deep-clone/upload-files',
    headers={'Authorization': os.environ['POWERUSER_KEY']},
    json={
        'deepVoiceCloneId': 'dV7kM3pQ9w',
        'audioS3Uris': [
            's3://autessa-voice/voice-clones/audio/prod/1234/8f3a2b.wav',
            's3://autessa-voice/voice-clones/audio/prod/1234/2c9d7e.wav'
        ]
    }
)

files_uploaded = response.json()['filesUploaded']

Response
{
  "filesUploaded": 2,
  "errors": []
}

Start a fine-tune

POST/voice/deep-clone/start-finetune

Kicks off fine-tuning on the clone's dataset and moves the clone to status TRAINING. Training is asynchronous — poll /deep-clone/get-status until it reports COMPLETED (or FAILED). If the fine-tune cannot be started, the clone is marked FAILED with the message in errors.

Requires WRITE or ADMIN access on the clone, and that the dataset exists.

Request parameters

Name
deepVoiceCloneId
Type
string
Description
The deep clone to train.

Response

Name
fineTuneJobId
Type
string
Description
The speech provider's fine-tune job identifier.
Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/deep-clone/start-finetune \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "deepVoiceCloneId": "dV7kM3pQ9w" }'

Response
{
  "fineTuneJobId": "ft_9c2e5b7a1d",
  "errors": []
}

Get fine-tune status

POST/voice/deep-clone/get-status

Returns the clone's training status. While the clone is still in progress (PENDING, UPLOADING, TRAINING) this endpoint live-polls the speech provider and updates the stored progress; once the clone is COMPLETED or FAILED the cached result is returned without polling. On completion, serviceVoiceId is populated and the voice becomes usable.

Requires any access to the clone.

Request parameters

Name
deepVoiceCloneId
Type
string
Description
The deep clone to check.

Response

Name
status
Type
string
Description
PENDING, UPLOADING, TRAINING, COMPLETED, or FAILED.
Name
progress
Type
integer
Description
Training progress, 0–100.
Name
serviceVoiceId
Type
string
Description
The speech provider's voice identifier. Populated once training completes; null before that.
Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
JavaScript
Python

curl https://api.autessa.com/autessa_poweruser/voice/deep-clone/get-status \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "deepVoiceCloneId": "dV7kM3pQ9w" }'

async function waitForClone(deepVoiceCloneId) {
  while (true) {
    const response = await fetch(
      'https://api.autessa.com/autessa_poweruser/voice/deep-clone/get-status',
      {
        method: 'POST',
        headers: {
          'Authorization': process.env.POWERUSER_KEY,
          'Content-Type': 'application/json'
        },
        body: JSON.stringify({ deepVoiceCloneId })
      }
    );
    const data = await response.json();
    if (data.status === 'COMPLETED' || data.status === 'FAILED') {
      return data;
    }
    await new Promise((r) => setTimeout(r, 30_000));
  }
}

import os
import time
import requests

def wait_for_clone(deep_voice_clone_id):
    while True:
        response = requests.post(
            'https://api.autessa.com/autessa_poweruser/voice/deep-clone/get-status',
            headers={'Authorization': os.environ['POWERUSER_KEY']},
            json={'deepVoiceCloneId': deep_voice_clone_id}
        )
        data = response.json()
        if data['status'] in ('COMPLETED', 'FAILED'):
            return data
        time.sleep(30)

Response
{
  "status": "TRAINING",
  "progress": 62,
  "serviceVoiceId": null,
  "errors": []
}

List all deep clones

POST/voice/deep-clone/get-all

Lists every deep clone visible to the linked user, grouped by how they have access.

Request parameters

None — send an empty object {}.

Response

Each entry is a summary with id, voiceName, serviceVoiceId, status, progress, sharingLevel, and createdAt.

Name
ownedClones
Type
array<DeepVoiceCloneSummary>
Description
Deep clones the linked user owns.
Name
companySharedClones
Type
array<DeepVoiceCloneSummary>
Description
Clones shared with the whole company by other users.
Name
sharedWithMeClones
Type
array<DeepVoiceCloneSummary>
Description
Clones shared specifically with the linked user.
Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/deep-clone/get-all \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{}'

Response
{
  "ownedClones": [
    {
      "id": "dV7kM3pQ9w",
      "voiceName": "Brand Voice",
      "serviceVoiceId": "b1f88932-527d-4b75-c78a-bf612e8e7182",
      "status": "COMPLETED",
      "progress": 100,
      "sharingLevel": "COMPANY",
      "createdAt": "2026-06-25T11:08:44"
    }
  ],
  "companySharedClones": [],
  "sharedWithMeClones": [],
  "errors": []
}

Get a deep clone

POST/voice/deep-clone/get

Fetches a single deep clone. This returns the same status payload as /deep-clone/get-status — status, progress, and serviceVoiceId — including the same live provider poll while training is in progress.

Requires any access to the clone.

Request parameters

Name
deepVoiceCloneId
Type
string
Description
The deep clone to fetch.

Response

Name
status
Type
string
Description
PENDING, UPLOADING, TRAINING, COMPLETED, or FAILED.
Name
progress
Type
integer
Description
Training progress, 0–100.
Name
serviceVoiceId
Type
string
Description
The speech provider's voice identifier, once training completes.
Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/deep-clone/get \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "deepVoiceCloneId": "dV7kM3pQ9w" }'

Response
{
  "status": "COMPLETED",
  "progress": 100,
  "serviceVoiceId": "b1f88932-527d-4b75-c78a-bf612e8e7182",
  "errors": []
}

POST/voice/deep-clone/share

Sets the clone's sharing level and grants per-user access — same semantics as sharing an instant clone: with SPECIFIC_MEMBER, userIds and permissionLevels are positionally matched arrays, defaulting to READ.

Requires ADMIN access on the clone.

Request parameters

Name
deepVoiceCloneId
Type
string
Description
The deep clone to share.
Name
shareLevel
Type
string
Description
PRIVATE, SPECIFIC_MEMBER, or COMPANY.
Name
companyPermissionLevel
Type
string
Description
When shareLevel is COMPANY, the access level every company member gets: READ, WRITE, PUBLISH, or ADMIN.
Name
userIds
Type
array<string>
Description
When shareLevel is SPECIFIC_MEMBER, the users to grant access to.
Name
permissionLevels
Type
array<string>
Description
Access level for each entry in userIds, positionally matched. Missing entries default to READ.

Response

Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/deep-clone/share \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "deepVoiceCloneId": "dV7kM3pQ9w",
    "shareLevel": "COMPANY",
    "companyPermissionLevel": "READ"
  }'

Response
{
  "errors": []
}

POST/voice/deep-clone/get-sharing-details

Returns the clone's current sharing configuration and the calling user's own access level. Requires any access to the clone.

Request parameters

Name
deepVoiceCloneId
Type
string
Description
The deep clone to inspect.

Response

Name
sharingLevel
Type
string
Description
PRIVATE, SPECIFIC_MEMBER, or COMPANY.
Name
companyPermissionLevel
Type
string
Description
Access level granted company-wide, when sharingLevel is COMPANY.
Name
sharedWith
Type
array<UserShare>
Description
Per-user grants: userEmail, userId (string), accessLevel.
Name
userAccessLevel
Type
string
Description
The calling user's effective access level.
Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/deep-clone/get-sharing-details \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "deepVoiceCloneId": "dV7kM3pQ9w" }'

Response
{
  "sharingLevel": "COMPANY",
  "companyPermissionLevel": "READ",
  "sharedWith": [],
  "userAccessLevel": "ADMIN",
  "errors": []
}

Transfer deep clone ownership

POST/voice/deep-clone/transfer-ownership

Transfers the clone to another member of the company. Requires ADMIN access on the clone.

Request parameters

Name
deepVoiceCloneId
Type
string
Description
The deep clone to transfer.
Name
newOwnerId
Type
string
Description
User ID of the new owner.

Response

Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/deep-clone/transfer-ownership \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "deepVoiceCloneId": "dV7kM3pQ9w",
    "newOwnerId": "uB7nT2wQ9k"
  }'

Response
{
  "errors": []
}

Delete a deep clone

POST/voice/deep-clone/delete

Permanently deletes a deep clone, whatever its training status. Agents configured to use the voice fall back to their built-in voice.

Requires ADMIN access on the clone.

Request parameters

Name
deepVoiceCloneId
Type
string
Description
The deep clone to delete.

Response

Name
errors
Type
array<string>
Description
Error messages, if any occurred.

cURL
curl https://api.autessa.com/autessa_poweruser/voice/deep-clone/delete \
  -X POST \
  -H "Authorization: $POWERUSER_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "deepVoiceCloneId": "dV7kM3pQ9w" }'

Response
{
  "errors": []
}

Errors

Status	When
`200`	Success — check `errors` for validation messages.
`400`	Feature gating or validation: voice cloning / fine-tuning not enabled for the company, blank `voiceName`, missing `audioS3Uri`, or dataset not created yet.
`401`	The linked user lacks the required access level (see the access table).
`403`	Scoped key without the endpoint in its allow-list.
`404`	Voice clone not found.
`408`	Unexpected server error; `errors` contains the message.
`500`	Speech-provider call failed (clone creation, dataset upload, fine-tune start, or status poll); `errors` contains the message.

Failures during /deep-clone/upload-files or /deep-clone/start-finetune also mark the clone FAILED, which is then reported by /deep-clone/get-status.

Instant clones vs deep clones​

Generate an audio upload link​

Request parameters​

Response​

Create an instant clone​

Request parameters​

Response​

Get an instant clone​

Request parameters​

Response​

List all instant clones​

Request parameters​

Response​

Delete an instant clone​

Request parameters​

Response​

Share an instant clone​

Request parameters​

Response​

Get instant clone sharing details​

Request parameters​

Response​

Transfer instant clone ownership​

Request parameters​

Response​

Create a deep clone dataset​

Request parameters​

Response​

Upload files to a dataset​

Request parameters​

Response​

Start a fine-tune​

Request parameters​

Response​

Get fine-tune status​

Request parameters​

Response​

List all deep clones​

Request parameters​

Response​

Get a deep clone​

Request parameters​

Response​

Share a deep clone​

Request parameters​

Response​

Get deep clone sharing details​

Request parameters​

Response​

Transfer deep clone ownership​

Request parameters​

Response​

Delete a deep clone​

Request parameters​

Response​

Errors​

Related​

Instant clones vs deep clones

Generate an audio upload link

Request parameters

Response

Create an instant clone

Request parameters

Response

Get an instant clone

Request parameters

Response

List all instant clones

Request parameters

Response

Delete an instant clone

Request parameters

Response

Share an instant clone

Request parameters

Response

Get instant clone sharing details

Request parameters

Response

Transfer instant clone ownership

Request parameters

Response

Create a deep clone dataset

Request parameters

Response

Upload files to a dataset

Request parameters

Response

Start a fine-tune

Request parameters

Response

Get fine-tune status

Request parameters

Response

List all deep clones

Request parameters

Response

Get a deep clone

Request parameters

Response

Share a deep clone

Request parameters

Response

Get deep clone sharing details

Request parameters

Response

Transfer deep clone ownership

Request parameters

Response

Delete a deep clone

Request parameters

Response

Errors

Related