Pinned Tweet
$52M seed round. 15M+ users. But this is what we're most proud of yet.
Introducing Drama 3: the most emotive and controllable TTS model in the world
And yes, it's open-weight.
You can now control how even a single word should sound, change its tone, and shift your voice ger was impressed too.
Try it here:
s.fish.audio/rqh0l8
Comment “DRAMA” within the next 48 hours for 1 month free.
If you're an enterprise your first month is on us:
s.fish.audio/qz9r0t
Introducing Drama 3: the most emotive and controllable TTS model in the world
And yes, it's open-weight.
You can now control how even a single word should sound, change its tone, and shift your voice ger was impressed too.
Try it here:
s.fish.audio/rqh0l8
Comment “DRAMA” within the next 48 hours for 1 month free.
If you're an enterprise your first month is on us:
s.fish.audio/qz9r0t
8,320
1,645
17,235
13,023,620
If you're building a game, we’d love to hear from you!
Let’s give your next cast a voice.
DMs are open 🐟
Let’s give your next cast a voice.
DMs are open 🐟
Porting Bully into Unreal Engine with Opus 5.5 PT3: AI NPCs
I gave every single NPC their own consciousness and ability to speak
I plugged in Fish Audio plus a few other fun things and you can now talk to any character in the game with the capability for them not only to ions but to speak back to you
You no longer have 10 important NPCs. You can now have 1,000 NPCs with unlimited options instead of just 8 dialogue prompts to choose from
Imagine talking to Patches from Dark Souls 1, but asking him questions about his travels or learning more lore about him after you speak to him for 10 minutes straight
Instead of just the same dialogue getting exhausted every time you speak to an NPC, you can now talk to him for as long as you want, and he'll respond to you in character
You can now literally just sit there and talk to them for hours if you really wanted to (if you wanted more ai psychosis lmaoo)
And if you plug in other AIs like Jev, you can give them the ability to play the game on their own, like how in Animal Crossing you can give someone a job and they'll get wood for you
Imagine the ability to talk to anybody and the capability of them doing anything in-game for you
(fully OS, free, never monetizing, needs a paid copy of Bully to play, ran on a macbook, link below, full youtube video soon)
Next: Full map port
9
2
54
3,405
One script. Unlimited emotions. 🎭
With Fish Audio Drama 3, you direct every performance, seriously!
Try it on our web app and API 🐳
With Fish Audio Drama 3, you direct every performance, seriously!
Try it on our web app and API 🐳
39
125
1,591
190,443
Spooky voices from Fish Audio, now on @livekit 🎃 Try them out and build something creepy this Halloween 👻
Meet SpookyKit, five haunted voices for your voice agent 🎃
Voices from LiveKit coworkers (guess who?), via @FishAudio as a custom voice:
🕯️ the hollow whisperer (v_rcMdmMksGhqr)
📞 the grave caller (v_ou7cbav4LMuq)
🌊 the thing below (v_U9DZ7HDMqqtM)
Voices summoned from the riends at @GradiumAI 🧡
💀 the bone barker (hwchrc8s07s20000)
📻 the dead air host (hwchrc8s07s00000)
paste a voice ID into LiveKit Inference or Agent Builder and go give someone a fright 👻
Try them now >
cloud.livekit.io
3
1
26
3,696
The fun begins 🐠 We can't wait to hear what y'all cook up!
Our proudest launch ever
We're here to make voice AI accessible to everyone.
Enjoy building with Drama 3!
x.com/i/trending/211…
13
1
65
4,167
Drama 3 Preview is the #1 open-weights TTS model on @ArtificialAnlys 🎭
Beats Eleven v4 on pronunciation. 5x cheaper. 83 languages.
Weights on HF, live on our API today.
Best-in-class voice AI, for everyone.
And we're just getting started.
Beats Eleven v4 on pronunciation. 5x cheaper. 83 languages.
Weights on HF, live on our API today.
Best-in-class voice AI, for everyone.
And we're just getting started.
Fish Audio's Drama 3 Preview is the new #1 open-weights Text to Speech model on the Artificial Analysis Text to Speech leaderboard. The model also leads all models in our Pronunciation Robustness leaderboard, ahead of Eleven v4 and Eleven v4 Turbo
Drama 3 Preview is the new open model from @FishAudio, built for multi-character scenes and long-form content with voice cloning and natural-language control. Weights are available on Hugging Face under the Fish Audio Community License.
Key takeaways:
➤ Pronunciation Robustness: Drama 3 Preview scores 93.2%, ranking #1 ahead of Eleven v4 at 91.7% and Eleven v4 Turbo at 90.1%. It ranks #1 in Standalone Terms and Preserving Exact Sequences and #2 in Expanding Shorthand.
➤ Provider Voice: Drama 3 Preview has an Elo of 1,243 (+16/-16) across 1,857 appearances, ranking #10, a 101 point gain on the previous Fish Audio S2.1 Pro at 1,142.
➤ Controlled Voice: Drama 3 Preview ranks #14 with an Elo of 1,058 (+15/-15) across 1,804 appearances.
➤ Open weights: Drama 3 Preview is the #1 open weights model on all three of our Text to Speech leaderboards. On Pronunciation Robustness it leads the next open weights model, Mistral's Voxtral TTS at 78.1%, by 15 points, and in the Provider Voice Arena it is ahead of Breeze TTS 2 at 1,221. Drama 3 is a 35B parameter mixture-of-experts model (3B active) supporting 83 languages, available on Hugging Face.
➤ Pricing: Drama 3 Preview is available via the Fish Audio API at $15/1M characters, the same as S2.1 Pro, compared to Eleven v4 at $80/1M characters and Eleven v4 Turbo at $40/1M characters.
See more details and listen to samples below 🧵
23
9
136
11,609
$52M seed round. 15M+ users. But this is what we're most proud of yet.
Introducing Drama 3: the most emotive and controllable TTS model in the world
And yes, it's open-weight.
You can now control how even a single word should sound, change its tone, and shift your voice ger was impressed too.
Try it here:
s.fish.audio/rqh0l8
Comment “DRAMA” within the next 48 hours for 1 month free.
If you're an enterprise your first month is on us:
s.fish.audio/qz9r0t
7
0
39
6,716
Fish Audio retweeted
Fish Audio's Drama 3 Preview is the new #1 open-weights Text to Speech model on the Artificial Analysis Text to Speech leaderboard. The model also leads all models in our Pronunciation Robustness leaderboard, ahead of Eleven v4 and Eleven v4 Turbo
Drama 3 Preview is the new open model from @FishAudio, built for multi-character scenes and long-form content with voice cloning and natural-language control. Weights are available on Hugging Face under the Fish Audio Community License.
Key takeaways:
➤ Pronunciation Robustness: Drama 3 Preview scores 93.2%, ranking #1 ahead of Eleven v4 at 91.7% and Eleven v4 Turbo at 90.1%. It ranks #1 in Standalone Terms and Preserving Exact Sequences and #2 in Expanding Shorthand.
➤ Provider Voice: Drama 3 Preview has an Elo of 1,243 (+16/-16) across 1,857 appearances, ranking #10, a 101 point gain on the previous Fish Audio S2.1 Pro at 1,142.
➤ Controlled Voice: Drama 3 Preview ranks #14 with an Elo of 1,058 (+15/-15) across 1,804 appearances.
➤ Open weights: Drama 3 Preview is the #1 open weights model on all three of our Text to Speech leaderboards. On Pronunciation Robustness it leads the next open weights model, Mistral's Voxtral TTS at 78.1%, by 15 points, and in the Provider Voice Arena it is ahead of Breeze TTS 2 at 1,221. Drama 3 is a 35B parameter mixture-of-experts model (3B active) supporting 83 languages, available on Hugging Face.
➤ Pricing: Drama 3 Preview is available via the Fish Audio API at $15/1M characters, the same as S2.1 Pro, compared to Eleven v4 at $80/1M characters and Eleven v4 Turbo at $40/1M characters.
See more details and listen to samples below 🧵
Drama 3 Preview is the new open model from @FishAudio, built for multi-character scenes and long-form content with voice cloning and natural-language control. Weights are available on Hugging Face under the Fish Audio Community License.
Key takeaways:
➤ Pronunciation Robustness: Drama 3 Preview scores 93.2%, ranking #1 ahead of Eleven v4 at 91.7% and Eleven v4 Turbo at 90.1%. It ranks #1 in Standalone Terms and Preserving Exact Sequences and #2 in Expanding Shorthand.
➤ Provider Voice: Drama 3 Preview has an Elo of 1,243 (+16/-16) across 1,857 appearances, ranking #10, a 101 point gain on the previous Fish Audio S2.1 Pro at 1,142.
➤ Controlled Voice: Drama 3 Preview ranks #14 with an Elo of 1,058 (+15/-15) across 1,804 appearances.
➤ Open weights: Drama 3 Preview is the #1 open weights model on all three of our Text to Speech leaderboards. On Pronunciation Robustness it leads the next open weights model, Mistral's Voxtral TTS at 78.1%, by 15 points, and in the Provider Voice Arena it is ahead of Breeze TTS 2 at 1,221. Drama 3 is a 35B parameter mixture-of-experts model (3B active) supporting 83 languages, available on Hugging Face.
➤ Pricing: Drama 3 Preview is available via the Fish Audio API at $15/1M characters, the same as S2.1 Pro, compared to Eleven v4 at $80/1M characters and Eleven v4 Turbo at $40/1M characters.
See more details and listen to samples below 🧵
15
40
630
59,253
We had a blast shooting this with @MrNigelNg!
$52M seed round. 15M+ users. But this is what we're most proud of yet.
Introducing Drama 3: the most emotive and controllable TTS model in the world
And yes, it's open-weight.
You can now control how even a single word should sound, change its tone, and shift your voice ger was impressed too.
Try it here:
s.fish.audio/rqh0l8
Comment “DRAMA” within the next 48 hours for 1 month free.
If you're an enterprise your first month is on us:
s.fish.audio/qz9r0t
18
1
158
16,367
Drama 3 is live on day 1 across @ComfyUI, @openart_ai, @runwayml, @Creatify_AI, @TapNow_AI, @astorie_ai, @pika_labs, @OpenRouter, @vercel, @OrcaRouter, @runware, @Inference_ai, @openclaw, Hermes agent (@NousResearch). Proud to have you all as our launch partners 🙏
10
72
596
471,530
And huge thanks to @sgl_project and @radixark for day-0 support of the Drama 3 open-weights model!
3
63
424
381,154
Add more emotions to your character’s voice.
Fix a line or even a single word without regenerating anything else.
Now in Drama 3. 🐳
Fix a line or even a single word without regenerating anything else.
Now in Drama 3. 🐳
18
12
133
12,670
For drama 3 preview API access: docs.fish.audio/api-reference/… and set the model to [drama-3-preview].
For web use: fish.audio/app/text-to-sp…. Choose Fish Audio Drama 3 (Preview) model.
For web use: fish.audio/app/text-to-sp…. Choose Fish Audio Drama 3 (Preview) model.
3
0
14
6,507
Duolingo teaches you what to say. This teaches you how to hold a conversation. @whosamberella built a language app for practicing real conversations with AI voices.
- Every character has their own personality
- Comes with pronunciation cards for beginners
- Characters even get get it
Every voice is AI and built with Drama 3.
- Every character has their own personality
- Comes with pronunciation cards for beginners
- Characters even get get it
Every voice is AI and built with Drama 3.
38
931
1,018
58,218
@whosamberella For drama 3 preview API access: docs.fish.audio/api-reference/… and set the model to [drama-3-preview]. Grab your API key at fish.audio/app/api-keys
For web use: fish.audio/app/text-to-sp…. Choose Fish Audio Drama 3 (Preview) model
For web use: fish.audio/app/text-to-sp…. Choose Fish Audio Drama 3 (Preview) model
0
0
8
7,685
Drama 3 doesn't just read lines. It brings the drama.
Direct the performance in plain language: tone, emotion, pacing, character. It can even shift emotion mid-line, no retake needed.
Try the preview now in the API with 'drama-3-preview'.
Direct the performance in plain language: tone, emotion, pacing, character. It can even shift emotion mid-line, no retake needed.
Try the preview now in the API with 'drama-3-preview'.
15
12
130
419,322
If you are a developer, open this doc: docs.fish.audio/api-reference/… and set the model to [drama-3-preview].
Grab your API key at fish.audio/app/api-keys
If you are a web user, ppen: fish.audio/app/text-to-sp…. Choose Fish Audio Drama 3 (Preview) model
Grab your API key at fish.audio/app/api-keys
If you are a web user, ppen: fish.audio/app/text-to-sp…. Choose Fish Audio Drama 3 (Preview) model
1
0
10
2,948
