Spanish from Spain, for real
Native voices from Spain with casting notes in Spanish, and native voices in Catalan, Galician and Basque. Plus 55+ languages with any voice.
- Castellano
- Català
- Galego
- Euskara
Voyager 1 speaks 55+ languages with 800+ voices, at 48 kHz, and signs every file with C2PA. Its API takes the ElevenLabs SDK: change the base URL and the key, and your integration keeps running.
Get your API keyclient = ElevenLabs( api_key="sk_…", api_key="vf_…", base_url="https://api.voicefinder.io",)Only facts you can check. Where they win, we say so too.
ElevenLabs details come from their public pages, reviewed on 21 October 2026. We do not compare prices: theirs change often; check them on their site. ElevenLabs is a registered trademark of its owner; Voicefinder is not affiliated with ElevenLabs.
Native voices from Spain with casting notes in Spanish, and native voices in Catalan, Galician and Basque. Plus 55+ languages with any voice.
Inaudible watermark and signed Content Credentials on every file, on every plan, in line with Article 50 of the EU AI Act. Anyone can check it.
Verify an audio fileEvery catalog voice has its licence and provenance documented, in the open. Your clone is private and comes with a signed rights statement.
Where our voices come fromWhen you need a person, on the same platform: post a casting and get proposals from professional voice actors in under 24 hours.
Find voice actorsThe official ElevenLabs SDK, in Python and JavaScript, works by changing the base URL and the key. eleven_* model_id values are accepted with no error.
API for developersFrom €0, with the API on every plan. 1 credit = 1 character: Plus, at €20 a month, is about 206 minutes of audio. Pay yearly and save 30%.
See pricingOn the same public Spanish test set, Voyager 1 gets fewer words wrong than ElevenLabs v4. We measure it with the standard method and publish it with its sources.
Not one word skipped, repeated or made up in 480 hand-checked takes: plain sentences in Spanish and English, and sentences with emotion tags.
MiniMax TTS-Multilingual-Test-Set (100 sentences, one take each), transcribed with Whisper-large-v3 and its normalisation. Competitors: MiniMax-Speech and ElevenLabs Multilingual v2, arXiv:2505.07916 (table 2); Fish Audio Drama 3 and ElevenLabs v4, arXiv:2610.18010 (table 6), where ElevenLabs v4 was measured through its API. Voyager 1: our own measurement on 21 Oct 2026 with the production model.
Every Voyager 1 measurementIf you need something on this list today, ElevenLabs has it. With a compatible API you can use both: them for that, Voicefinder for the voice.
Three steps and the same SDK. What changes is marked in yellow.
https://api.voicefinder.ioNo /v1: the SDK adds it.
vf_…Create it in Developers. Send it in xi-api-key, x-api-key or Authorization: Bearer.
client.voices.search()voice_id values come from our catalog. Your ElevenLabs voices do not carry over: clone your own from a recording.
from elevenlabs.client import ElevenLabs client = ElevenLabs( api_key="sk_…", api_key="vf_…", base_url="https://api.voicefinder.io",) audio = client.text_to_speech.convert( voice_id="f6abdab0-2524-43d1-92e2-b39486f39fbe", # Agatha text="[upbeat] Hi! You're now talking to Voicefinder.", model_id="eleven_multilingual_v2", language_code="en",) with open("hello.mp3", "wb") as f: for chunk in audio: f.write(chunk)For text to speech, yes: Voyager 1 has 800+ voices in 55+ languages, 48 kHz audio and, on the public Spanish test set, a lower word error rate than ElevenLabs v4 (0.90% vs 1.09%). If you also need speech to text, a voice changer or music from a prompt, today only ElevenLabs has those.
Yes. The official ElevenLabs SDK, in Python and JavaScript, works by changing the base URL to https://api.voicefinder.io and the key to a Voicefinder one (vf_). The API accepts their field names, headers and formats, and eleven_* model_id values are treated as voyager-1 with no error.
No: voice_id values come from our catalog, which you search with voices.search(). If you want your own voice, clone it from a 10 to 15 second recording: it stays private to your account, with a signed rights statement.
Listen: the demos on this page come from the catalog. In numbers, Voyager 1 scores 0.90% word error in Spanish and 1.53% in English, with 0 failed sentences in 480 hand-checked takes. All at 48 kHz.
Speech to text, a voice changer, a voice isolator, music and voice design from a prompt, WebSocket streaming (ours arrives with Pulsar 1), certifications such as SOC 2 or ISO 27001, and a choice of data residency. We say it plainly so you can choose well.
From €0 a month (personal use). Basic is €11, Plus €20, Pro €89, Growth €269 and Business €809 a month; Enterprise from €899. Pay yearly and save 30%. 1 credit = 1 character and the API is on every plan.
No. ElevenLabs is a registered trademark of its owner and Voicefinder is not affiliated with or endorsed by it. We name it to compare and to explain API compatibility.
Create your free account and hear Voyager 1. If you already have code, change two lines.