The voice layer for agents. Real-time, accurate, natural voice APIs: TTS, STT, turn-taking & voice design. Devs: build → gradium.ai
- Tweets 234
- Following 19
- Followers 4,466
- Likes 384
Pinned Tweet
Today, we’re launching Voice Design.
Write a prompt, create a voice.
Describe the accent, age, gender, and pace your use case needs, and get new voices in seconds, ready to use.
Live and free in the Gradium API and Studio. gradium.link/voice-design
79
134
656
443,659
Welcome to our weekly community spotlight, featuring the top three teams from the AI Gaming Hack in Paris, all built with Gradium.
Here’s how each team used Gradium in their winning game 📷
Explore their projects: gradium.link/project-gallery
1
4
19
3,224
2nd Place - Last Alibi
Investigate a murder aboard a snowbound 1931 train by questioning suspects aloud.
Gradium Speech-to-Text transcribes your questions, and Text-to-Speech gives the suspects their voices.
last-alibi.kaisspace.workers…
1
2
143
3rd Place - AfterFlow
Ferry souls through the underworld and read scroll incantations aloud to calm the gods.
The team used Gradium Speech-to-Text for incantations and Text-to-Speech to create the gods’ voice lines.
afterflow.stream
3
77
Our default Text-to-Speech model now has around 50ms time to first audio.
The model scores highest on naturalness for any sub-100ms model on Speko public benchmarks.
How much of your turn budget does TTS take?
gradium.link/fastest-tts
8
9
46
6,630
Gradium will be at Startup Day France tomorrow at @joinstationf Paris. Join us at the founder panel during the 14:00–15:00 CEST plenary.
cloud.google.com/events/intl…
2
19
1,646
Our first build with Jev from @typesafeai: a support agent that design its voice from the call context.
What would you build?
gradium.ai/blog/voice-that-a…
1
3
17
883
Thank you to all who joined us at the AI Gaming Hackathon in Paris.
Check out the projects at gradium.link/project-gallery
See you at the next hackathon.
4
14
1,717
Our community page is live.
See apps built with Gradium, share yours in the showcase, and talk to our team on Discord.
Hear about upcoming hackathons and events.
Selected projects will be featured in the gallery and qualify for bonus credits.
gradium.ai/community
4
11
614
Gradium's TTS beta model offers the best combination of naturalness and speed on @speko_ai's new benchmark. Try it now: studio.gradium.ai/?model_nam…
5
18
905
Join us in Paris and Zurich for the upcoming AI Gaming and Agentic AI Hack.
Every participant gets free Gradium credits to add voice to what they build. Registration links are below.
luma.com/par-hack
1
1
13
1,955
2/3 Sep 26 · AI Gaming Hack, Paris
Build the best gaming experience possible in 8 hours.
Register: luma.com/par-hack
1
4
247
3/3 Oct 10 · Agentic AI Hack, Zurich
Build an agent people can talk to within a day using Gradium.
Register: luma.com/zrh-hack
2
124
Gradium retweeted
We measured Gradium TTS Beta on our TTS board: 49ms p50, 56ms p90 time to first audio from US East.
That is more than 4x faster than the previous Gradium TTS (217ms p50), with the same naturalness.
Full results: benchmarks.speko.ai/tts
Gradium TTS beta model is out now to define what fast means.
Less than 50ms TTFA, same quality. Most models are >100ms TTFA.
Try it now: studio.gradium.ai/?model_nam…
7
25
3,384
Gradium TTS beta model is out now to define what fast means.
Less than 50ms TTFA, same quality. Most models are >100ms TTFA.
Try it now: studio.gradium.ai/?model_nam…
9
11
91
24,538
September 23:
Our Chief Growth Officer @ConstanceGriso joins Monetizing AI Agents with The Reasoning Project, Metronome, and Dust.
luma.com/monetizing-agents
1
2
5
298
September 24:
Our CEO @neilzegh Zeghidour will speak at HumanX.
humanx.co/europe
Our CTO @olivierteboul will speak at the AI Engineer Conference at STATION F.
ai.engineer/paris/2026
Our Chief Science Officer @honualx will speak at the Artificial Intelligence Marseille conference.
aim-marseille.com/en
2
7
366
With our voice designer you can generate voices that fit the character, and then perform the lip-sync. Try it with studio.gradium.ai
H3 Max Lip Sync is now live on fal.
Upload one photo and one audio clip and get back a lip-synced video in seconds, with natural expression and movement, in any language.
In our evals, H3 Max Lip Sync ranks #1 for both quality and speed, with a median generation time of just 11 seconds.
Try it now on fal.
Made with AI
6
3
19
1,131