Skip to main content
Version: 2.0.0

Changelog

May 1, 2024

Enhancements

Feb 12, 2024

Additions

  • Source information about a given Voice is now available in the source field of a request to GET /voices or GET /voices/:uuid. The source indicates the origin of the voice, for example: Custom Voice, Resemble Voice or Marketplace

Jan 26, 2024

Additions

  • Advanced information about Voice model versions is now available. By setting the advanced query parameter to true, information about the various models used for Voice features will be displayed
GET /voices?advanced=true
GET /voices/<voice_uuid>?advanced=true

{
// ... snipped
"model": {
"text_to_speech": "tts-v3",
"voice_conversion": "sts-v2",
"fill": "fill-v1",
}
}


Nov 7, 2023

Type Definitions Update

  • The @resemble/node version 3.3.2 is now available. This patch includes an update to the clip stream function's TypeScript definition. The sample_rate and precision properties were always part of the intended interface but were missing in the TypeScript type definitions provided in the previous release. This update corrects the oversight.

Nov 7 2023

Changes

  • The @resemble/node version 3.3.1 is now available. This patch includes an update to the clip stream function. The second parameter which you can use to specify buffer size, ignoring wav headers etc. within the stream function was intended to be optional in the previous release but was incorrectly marked as required. The updated function now correctly allows the omission of stream config object.

Nov 7 2023

Additions

  • The @resemble/node version 3.3.0 is now available, with support for the Phoneme and Term Substitution APIs. See the Node Phoneme API docs and the Node Term Substitution API docs for more information. Please be advised that version 3.2.0 of @resemble/node was unpublished from npm and is not available for download. As there were no downloads recorded for this version, it should not impact any existing installations.

Nov 6 2023

Additions

Nov 2 2023

Additions

Nov 1 2023

Additions

  • Synchronous and streaming requests are now out of beta! All users can access

the clip streaming and synchronous endpoints without beta permissions.

Aug 25 2023

Additions

  • You can now search for clips fetched through the /v2/projects/:project_uuid/clips endpoint based on title using the new query parameter q
GET /v2/projects/:project_uuid/clips?q=dead%20souls

{
"success": true,
"page": 1,
"num_pages": 1,
"page_size": 11,
"items": [
{
"uuid": "5a16c5df",
"title": "dead souls",
"body": "This was not the old Chichikov. This was some wreckage of the old Chichikov. The inner state of his soul might be compared to a demolished building, which has been demolished so that from it a new one could be built; but the new one has not been started yet, because the infinitive plan has not yet come from the architect and the workers are left in perplexity.",
"is_public": true,
"is_archived": false,
"created_at": "2023-08-25T16:56:13.871Z"
}
]
}

Removals

  • None

Jul 21 2023

Additions

  • You can now see the signal-to-noise, loudness and resemble sample scores metrics in the metrics field associated with a recording fetched through the /v2/voices/:voice_uuid/recordings endpoint.
  • See the documentation for more information
{
"success": true,
"item": {
"uuid": <string>,
"name": <string>,
"text": <string>,
"emotion": <string>,
"is_active": <boolean>,
"fill": <boolean>,
"audio_src": Optional<string>,
"created_at": <UTC string>,
"updated_at": <UTC string>,
"metrics": Optional<{
"signal_to_noise_ratio": <number>,
"loudness_grade": <string>,
"resemble_sample_score": <number>,
}>,
}
}

Removals

  • None

Jul 18 2023

Additions

  • We have made account management and multi-team usage easier with the addition of the /v2/account API.
  • You can now fetch individual user account information via /v2/account, or team information via /v2/account/teams/:uuid/
  • See the documentation for more information
{
"success": true,
"item": {
"uuid": "abcdef123",
"name": "a team",
"plan": "Trial",
"voice_limit": 10,
"units": "seconds",
"rate": 0.0,
"current_usage": 0
}
}

Removals

  • None

Jul 11 2023

Additions

  • We have made it easier to understand the current training state of your voice's fill, speech to speech, and text to speech capabilities with a new field called component_status.
    {
"uuid": <string>,
"name": <string>,
"status": <string>,
"default_language": <string>,
"supported_languages": <string[]>,
"component_status": {
"text_to_speech": { "status": <string> },
"speech_to_speech": { "status": <string> },
"fill": { "status": <string> }
},
}