AI-generated Answer for harlowregular123_21060
Short answer: mostly fine for your app’s UI, but several fields don’t match ElevenLabs’ official voice schema or endpoints.
What the ElevenLabs API actually expects/returns
- Voice objects include fields such as voice_id, name, category, labels, description, preview_url, settings, etc. There is no avatar_url, provider, or preview_audio_url field (the correct field is preview_url). (elevenlabs.io)
- category is used to classify voices (e.g., premade, cloned, generated, professional). “custom” is not a documented category. (elevenlabs.io)
- voice_type is not a property on a voice; it’s only a filter parameter on the List voices endpoint (allowed values: personal, community, default, workspace, non-default). (elevenlabs.io)
How to validate and use this voice_id
- Verify the voice exists and fetch its official fields (including preview_url) via GET /v1/voices/{voice_id}. (elevenlabs.io)
- Generate audio by passing the voice_id to Text to Speech:
- Batch: POST /v1/text-to-speech/{voice_id} (Convert). (elevenlabs.io)
- Realtime streaming: GET wss://api.elevenlabs.io/v1/text-to-speech/{voice_id}/stream-input. (elevenlabs.io)
Recommended mapping for your object
- voice_id: keep as-is (required by TTS). (elevenlabs.io)
- voice_name → name (matches API field). (elevenlabs.io)
- preview_audio_url → preview_url (matches API field). (elevenlabs.io)
- voice_type: remove or use only for your app; not part of the voice object. If you need filtering, use the List voices voice_type query param. (elevenlabs.io)
- provider, avatar_url: app-specific; not used by ElevenLabs APIs.
odd jasper
narrow ermine