D-ID taps ElevenLabs to give AI avatars better voices

The quickest way to ruin an AI avatar video is the voice. The face might look convincing enough, but the moment it speaks like a robotic GPS navigation system, the whole illusion falls apart.

Generative video platform D-ID announced a partnership with AI voice pioneer ElevenLabs, integrating ElevenLabs synthetic voices directly into D-ID's Creative Reality studio.

The update solves a long-standing workflow headache. Creators will no longer need to generate audio in ElevenLabs, download the file, and upload it back into D-ID just to sync lips.

Giving avatars an emotional voice

The integration is rolled out to D-ID Pro subscribers, allowing them to apply synthetic speech to uploaded photos, AI faces, or premium presenters with a single click.

D-ID is adding nine of ElevenLabs premium voices to its library. These voices are designed to deliver natural pacing, realistic emphasis, and richer personality.

The timing fits well with D-ID's recent interface upgrade, which introduced avatar facial expressions covering serious, happy, surprised, and neutral tones.

Pairing dynamic facial moods with ElevenLabs nuanced delivery should make virtual narrators feel noticeably less robotic.

Connecting the fragmented AI stack

Until now, hundreds of thousands of creators were already hacking these two tools together manually. Official integration simply legitimizes what users were already doing.

D-ID CEO Gil Perry noted that the studio is adding a new subscriber every two seconds on average, with over 150 million videos generated on the platform to date.

Founded in 2017, D-ID supports video narration in 119 languages and has already embedded its tools into software like Canva and Microsoft PowerPoint.

Meanwhile, ElevenLabs, founded in 2022 by Mati Staniszewski and Piotr Dabkowski, continues to solidify its spot as the go-to voice layer across the AI ecosystem.

Why it matters

We see this as a practical leap for corporate training teams, marketers, and solo creators churning out video content at scale.

If you produce social clips or training presentations, you no longer have to pay the productivity tax of bouncing between separate tabs. The friction drops significantly, making it well worth testing if you already hold a D-ID Pro plan.

Strategically, this partnership shows where generative AI is heading: consolidation over isolation. Point solutions that only do text-to-speech or only animate still frames are fragile on their own.

Instead of wasting millions reinventing each other's core models, D-ID and ElevenLabs teamed up to lock in enterprise and creator workflows. It makes life much harder for standalone competitors.

There is, however, an obvious downside. Lowering the bar to produce ultra-realistic talking heads with expressive audio also lowers the barrier for convincing disinformation. As these combined pipelines get faster and cheaper, the pressure on both teams to enforce ethical guardrails will intensify.