Video automation Video API Convert JSON to video Text-to-speech voices
Azure voice

ValentinaUruguay female neural voice

Valentina is a female neural voice from Microsoft Azure for Spanish (Uruguay). It reads at about 150 words per minute.

Voiceovers free to try · no credit card
Sample by Microsoft Azure
Free account, no credit card. Up to 60 seconds on the free plan.
#33of 75 Spanish Azure voices
NeuralGenerally available
150words per minute
—speaking styles

Similar voices

Spanish female voices with a close tone, from ElevenLabs and Azure.

Paloma
United States female neural voice
Spanish (United States)Female
#9 of 75 Spanish Azure
Maria
Costa Rica female neural voice
Spanish (Costa Rica)Female
#19 of 75 Spanish Azure
Karina
Puerto Rico female neural voice
Spanish (Puerto Rico)Female
#20 of 75 Spanish Azure
Nina
Expressive, Crisp, Dynamic
SpanishFemaleYoung
#453 of 1,660 Spanish
Silvia
Clear and Assertive
Spanish · PeninsularFemaleMiddle aged
#318 of 846 Spanish ElevenLabs
Make your voiceover with Valentina

Type your script and get the voiceover in about a minute. Free account, no credit card.

Create voiceover — free

About Valentina

Valentina is a female neural voice from Microsoft Azure for Spanish (Uruguay). It reads at about 150 words per minute.

Valentina ranks #33 of 75 Spanish Azure voices by search interest in our gallery.

Provider
Azure
Language
Spanish (Uruguay)
Gender
Female
Type
Neural · GA
Speed
150 words/min
Voice ID
es-UY-ValentinaNeural

Questions about Valentina

Can I make a voiceover with Valentina for free?

You can try it with a free JSON2Video account, no credit card: free voiceovers are up to 60 seconds. Paid plans remove the limit.

Can I use my Valentina voiceovers on YouTube or in ads?

Yes. You can publish the voiceovers and videos you make with Valentina on YouTube, TikTok, podcasts and ads, and use them in commercial projects.

Which languages can Valentina speak?

Valentina is made for Spanish (Uruguay). For other languages, pick a multilingual voice.

Does Valentina support SSML?

Yes. Use SSML to add pauses, change the speed or pitch, spell words out.

For developers: use Valentina with the API

Add a voice element to any JSON2Video movie. It works the same from n8n, Make, Zapier and the CLI.

{
    "type": "voice",
    "model": "azure",
    "voice": "es-UY-ValentinaNeural",
    "text": "En primavera, el jardín cobra vida con flores coloridas y pájaros cantando. El viejo roble proporciona sombra para los visitantes, mientras las mariposas bailan entre las rosas. Una pequeña fuente crea sonidos de paz, convirtiéndose en el lugar perfecto para relajarse y disfrutar de la belleza de la naturaleza."
}

Voice element reference

Browse more voices