Text-to-speech
Articles about Text-to-speech.
Eleven v4: What ElevenLabs' New Voice Model Changes for AI Ad Voiceovers
ElevenLabs launched Eleven v4 and a low-latency Eleven v4 Turbo on September 28, 2026. Here is what audio tags, two-speaker scenes, 90+ languages and 10-second clones change for a 30-second ad read, what a take costs on fal, and a test an ad team can run in one afternoon.
Gemini 3.8 Flash TTS for Ad Voiceovers: 30-Second Voice Replication, Consent First
Google's September 23 launch adds prompt-designed voices, a library it puts at 2,000-plus voices, and replication of your own voice or one you have the rights to use from a short sample, gated by a spoken consent recording from the voice owner. Here is what that means for brand voices and spokesperson ads, and what a read costs.
The New Top-Ranked Text-to-Speech Model Ships Two Voices. What Ad Voiceovers Actually Need
Alibaba's new speech model took the number one spot on Artificial Analysis' Speech Arena for provider voices at an Elo of 1,238, yet its published catalog is two system voices in Mandarin and English with Spanish still missing from the shipping language table, which is exactly the distance between winning a blind listening test and being usable for a Spanish-language ad.