Fish Audio(@FishAudio)

If you are a developer, open this doc: https://t.co/gBHOjYmlzV and set the model to [drama-3-preview...

Fish Audio (@FishAudio) — Wed, 07 Oct 2026 17:00:01 GMT

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

Your browser does not support the video tag.

🔗 View on Twitter

💬0🔄0❤️2👀290📊1

⚡ Powered by xgo.ing

Drama 3 doesn't just read lines. It brings the drama. Direct the performance in plain language: ton...

Fish Audio (@FishAudio) — Wed, 07 Oct 2026 17:00:00 GMT

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'.

Your browser does not support the video tag.

🔗 View on Twitter

💬7🔄10❤️42👀2634📊19

⚡ Powered by xgo.ing

Claude Sonnet 5.5 + Fish Audio MCP = 🤌 See prompts in the thread.

Fish Audio (@FishAudio) — Thu, 01 Oct 2026 17:04:05 GMT

Claude Sonnet 5.5 + Fish Audio MCP = 🤌 
See prompts in the thread.

Fish Creative@FishCreativeHQ

Claude Sonnet 5.5 is scary good at art direction!

One poster → a 10-second film, title and sound included.

All in one chat, with Fish Audio MCP image+video generations. See prompts ↓

Your browser does not support the video tag.

🔗 View on Twitter

🔗 View Quoted Tweet

💬6🔄3❤️17👀2580📊8

⚡ Powered by xgo.ing

Today, we're ending boring transcripts. Fish Audio’s ASR model now: → Identifies different speakers ...

Fish Audio (@FishAudio) — Mon, 28 Sep 2026 17:00:00 GMT

Today, we're ending boring transcripts. Fish Audio’s ASR model now:
→ Identifies different speakers
→ Understands what they're feeling
→ Labels emotion cues like [laughter], [surprised] in the transcript

All this while supporting 83 languages and being the most accurate STT model out there.

Try it out now 🐟

Your browser does not support the video tag.

🔗 View on Twitter

💬22🔄47❤️634👀43974📊210

⚡ Powered by xgo.ing

https://t.co/R70ttrHrPj

Fish Audio (@FishAudio) — Mon, 28 Sep 2026 17:00:00 GMT

docs.fish.audio/features/speec…

💬1🔄0❤️14👀3191📊6

⚡ Powered by xgo.ing

Welcome to the Fish Audio Startup Program! 3D characters have entered the chat 🐳 💬

Fish Audio (@FishAudio) — Sun, 27 Sep 2026 17:05:47 GMT

Welcome to the Fish Audio Startup Program! 
3D characters have entered the chat 🐳 💬

Hydrilla AI@HydrillaAI

Our 3D characters are about to find their voice 🎙️

We’re joining the @FishAudio Startup Program! Our next chapter brings production-ready 3D and expressive voice AI closer together.

Thanks to the Fish Audio team for backing us. We’re excited to share what’s next.

🔗 View Quoted Tweet

💬3🔄1❤️11👀2486📊4

⚡ Powered by xgo.ing

Check out our Startup Program: https://t.co/Bn27bXMKbG

Fish Audio (@FishAudio) — Sun, 27 Sep 2026 17:05:47 GMT

Check out our Startup Program:  fish.audio/startup/

💬1🔄0❤️3👀973📊2

⚡ Powered by xgo.ing

Start building with Drama 3 (preview): https://t.co/gBHOjYmTpt Route the model to “drama-3-preview”

Fish Audio (@FishAudio) — Fri, 25 Sep 2026 02:32:11 GMT

Start building with Drama 3 (preview): docs.fish.audio/api-reference/… 
Route the model to “drama-3-preview”

💬0🔄0❤️0👀868

⚡ Powered by xgo.ing

Introducing Drama 3 (preview): the most controllable TTS model ever. Describe the tone, pacing, and...

Fish Audio (@FishAudio) — Wed, 23 Sep 2026 17:01:51 GMT

Introducing Drama 3 (preview): the most controllable TTS model ever.

Describe the tone, pacing, and character in simple language, without thinking in audio tags. 

Shift your voice mid-sentence, generate a multi-character scene, or fix only a single word. 

Comment DRAMA to get an API key 🐟

Your browser does not support the video tag.

🔗 View on Twitter

💬1292🔄150❤️1939👀160949📊1847

⚡ Powered by xgo.ing

Connect your agent: https://t.co/Keue0KEt9O

Fish Audio (@FishAudio) — Tue, 22 Sep 2026 17:00:01 GMT

Connect your agent: docs.fish.audio/overview/mcp

💬0🔄0❤️8👀1470📊2

⚡ Powered by xgo.ing

Fish Audio MCP just got a major upgrade with image + video generation. 🎬 Turn a brief into an ad...

Fish Audio (@FishAudio) — Tue, 22 Sep 2026 17:00:00 GMT

Fish Audio MCP just got a major upgrade with image + video generation. 🎬  

Turn a brief into an ad without leaving your agent: generate visuals → make edits → bring them to life as videos. And see the cost before you create.

Claude. Codex. Cursor. Windsurf. 
Now make yours. 🎣

Your browser does not support the video tag.

🔗 View on Twitter

💬28🔄8❤️101👀7430📊50

⚡ Powered by xgo.ing

Fish Audio S2.1 Pro is now on @runwayml 🤘 This season, give your Fall'26 collection an expressive v...

Fish Audio (@FishAudio) — Mon, 21 Sep 2026 14:52:57 GMT

Fish Audio S2.1 Pro is now on @runwayml 🤘
This season, give your Fall'26 collection an expressive voice.

Runway@runwayml

The Runway Fall 2026 collection has landed.

Our biggest model drop yet, featuring Fish Audio S2.1 Pro, MiniMax H3 Max, Cartesia Sonic 3.6, Flux Video upscaler and edit, now available alongside the best models from Runway and other frontier labs. Get started at the link below.

Your browser does not support the video tag.

🔗 View on Twitter

🔗 View Quoted Tweet

💬3🔄2❤️21👀3515📊6

⚡ Powered by xgo.ing

Live speech-to-speech translation, powered by Fish Audio. 🐟 Built with our streaming TTS, this dem...

Fish Audio (@FishAudio) — Fri, 18 Sep 2026 17:00:01 GMT

Live speech-to-speech translation, powered by Fish Audio. 🐟

Built with our streaming TTS, this demo supports one-way translation and two-way conversations in 80+ languages.

Try it, explore the implementation, and start building your own ↓

Your browser does not support the video tag.

🔗 View on Twitter

💬4🔄1❤️30👀3642📊12

⚡ Powered by xgo.ing

https://t.co/clgX1nr17K by @cshape_

Fish Audio (@FishAudio) — Fri, 18 Sep 2026 17:00:01 GMT

translate.demo.fish by @cshape_

💬1🔄0❤️1👀986📊1

⚡ Powered by xgo.ing

How live translation works: https://t.co/9IxtKBiLok

Fish Audio (@FishAudio) — Fri, 18 Sep 2026 17:00:01 GMT

How live translation works: translate.demo.fish/how-it-works

💬0🔄0❤️0👀349

⚡ Powered by xgo.ing

Limited seats. Apply now👇 https://t.co/kWoWEHM9A8

Fish Audio (@FishAudio) — Thu, 17 Sep 2026 17:00:00 GMT

Limited seats. Apply now👇luma.com/tegocn0h8

💬0🔄0❤️2👀865

⚡ Powered by xgo.ing

🐟 We’re bringing back our Voice AI Builder Dinner with @covaldev in SF. Join other founders and p...

Fish Audio (@FishAudio) — Thu, 17 Sep 2026 17:00:00 GMT

🐟 We’re bringing back our Voice AI Builder Dinner with @covaldev in SF. 

Join other founders and product leaders on Sept 22!

We'll be exchanging insights on the hardest parts of building voice AI for healthcare, customer support, and other reliability-critical use cases.🎙️

💬2🔄1❤️8👀1691📊3

⚡ Powered by xgo.ing

Apply here: https://t.co/6QO84MUzMv

Fish Audio (@FishAudio) — Wed, 16 Sep 2026 17:00:02 GMT

Apply here: livekit.com/startups

💬0🔄0❤️2👀850📊1

⚡ Powered by xgo.ing

Fish Audio is now part of the @LiveKit Startup Program 🚀 Pre-Series A voice AI startups get: → $1,...

Fish Audio (@FishAudio) — Wed, 16 Sep 2026 17:00:01 GMT

Fish Audio is now part of the @LiveKit Startup Program 🚀

Pre-Series A voice AI startups get:
→ $1,000 in free Fish Audio credits
→ $22K+ more from LiveKit and other partners

Apply via link below ↓

💬8🔄2❤️41👀3447📊17

⚡ Powered by xgo.ing

Wan 3.0 now live on @FishCreativeHQ. 🐟

Fish Audio (@FishAudio) — Sat, 12 Sep 2026 02:36:01 GMT

Wan 3.0 now live on @FishCreativeHQ. 🐟

Fish Creative@FishCreativeHQ

The US Open has us thinking about our next shot. 🎾
Wan 3.0 is now live on Fish Creative.

79% lower cost than Seedance 2.5 → More room to try different angles and find the shot you love.

Your browser does not support the video tag.

🔗 View on Twitter

🔗 View Quoted Tweet

💬1🔄1❤️11👀2953📊3

⚡ Powered by xgo.ing

Better voice models start with better data. Our CEO @rissa_cao breaks down how we annotate and ev...

Fish Audio (@FishAudio) — Thu, 10 Sep 2026 17:00:01 GMT

Better voice models start with better data.  

Our CEO @rissa_cao breaks down how we annotate and evaluate emotional nuance, so models can preserve shifts in tone and delivery as a conversation unfolds.  

Expressive voice isn’t just about emotion. It’s about getting the right emotion at the right moment. 🐳

Your browser does not support the video tag.

🔗 View on Twitter

💬2🔄4❤️33👀3631📊11

⚡ Powered by xgo.ing

Try yourself ↓ https://t.co/l3wJgxD5gW

Fish Audio (@FishAudio) — Thu, 10 Sep 2026 17:00:01 GMT

Try yourself ↓
fish.audio/app/

💬0🔄0❤️3👀823📊1

⚡ Powered by xgo.ing

Tokyo 🇯🇵, you might see Fish audio on your commute this month 🐟 We’re now live at Shinagawa, Eb...

Fish Audio (@FishAudio) — Wed, 09 Sep 2026 16:00:00 GMT

Tokyo 🇯🇵, you might see Fish audio on your commute this month 🐟 

We’re now live at Shinagawa, Ebisu, and Toranomon Hills.

Your browser does not support the video tag.

🔗 View on Twitter

💬1🔄2❤️12👀1459📊4

⚡ Powered by xgo.ing

A resume only tells us what you've done. A project shows us how you build. So before anyone joins F...

Fish Audio (@FishAudio) — Thu, 03 Sep 2026 15:00:03 GMT

A resume only tells us what you've done. A project shows us how you build.

So before anyone joins Fish Audio full-time, we build something real together.

→ Strongly opinionated, weakly held
→ Solutions, not just questions
→ End-to-end ownership
→ Care deeply about the product

Our CEO @rissa_cao on what we actually look for, with CXFoundation. We're hiring 🐠

Your browser does not support the video tag.

🔗 View on Twitter

💬7🔄3❤️38👀3500📊15

⚡ Powered by xgo.ing

Open roles: → Voice Experience Designer → Enterprise Sales → Content / Performance Marketing → Deman...

Fish Audio (@FishAudio) — Thu, 03 Sep 2026 15:00:03 GMT

Open roles:
→ Voice Experience Designer
→ Enterprise Sales
→ Content / Performance Marketing
→ Demand Gen
Ship us something magical. DMs are open.

💬4🔄0❤️7👀1243📊5

⚡ Powered by xgo.ing

Full conversation: https://t.co/nWqgUOIAOZ

Fish Audio (@FishAudio) — Thu, 03 Sep 2026 15:00:03 GMT

Full conversation: youtube.com/watch?v=82l4A-…

💬0🔄0❤️3👀610📊1

⚡ Powered by xgo.ing

Fish Audio S2.1 Pro is now live in Europe and Australia through @slng_ai 83 languages. In-region G...

Fish Audio (@FishAudio) — Tue, 01 Sep 2026 17:00:00 GMT

Fish Audio S2.1 Pro is now live in Europe and Australia through @slng_ai 

83 languages. In-region GPU infrastructure. Closer to users, with the latency and enterprise controls production voice AI requires. 

Learn more: slng.ai/posts/fish-aud…

💬5🔄5❤️35👀2724📊12

⚡ Powered by xgo.ing

Download here👇 https://t.co/Gv9YX4u0qe

Fish Audio (@FishAudio) — Mon, 31 Aug 2026 18:34:55 GMT

Download here👇play.google.com/store/apps/det…e

💬0🔄0❤️8👀1276📊1

⚡ Powered by xgo.ing

"Android when?" → Android now 🐳 2M+ voices, 80+ languages, emotion you direct with open-ended tags...

Fish Audio (@FishAudio) — Mon, 31 Aug 2026 18:34:54 GMT

"Android when?" → Android now 🐳

2M+ voices, 80+ languages, emotion you direct with open-ended tags, and a voice clone from 5 seconds of audio.

Get it now ↓

Your browser does not support the video tag.

🔗 View on Twitter

💬7🔄5❤️114👀9366📊32

⚡ Powered by xgo.ing

Android, your turn. The whole voice studio.🐠 Get it at https://t.co/Gv9YX4u0qe.

Fish Audio (@FishAudio) — Fri, 28 Aug 2026 15:34:58 GMT

Android, your turn. The whole voice studio.🐠 
Get it atplay.google.com/store/apps/det…e.

Fish Audio@FishAudio

Today, Fish Audio goes mobile. 
Introducing Fish Audio iOS, the full voice studio, from desktop to your pocket.

🐳Discover 2M+ voices 
🔽Create multi-speaker dialogue with 80+ languages 
🔽Direct emotion with open-ended tags 
🔽Clone your own voice, or design one from a single prompt.

Try it now.

Your browser does not support the video tag.

🔗 View on Twitter

🔗 View Quoted Tweet

💬0🔄1❤️26👀4546📊5

⚡ Powered by xgo.ing

https://t.co/n9ddlrBfZA

Fish Audio (@FishAudio) — Thu, 27 Aug 2026 19:19:03 GMT

x.com/i/article/2093…

💬24🔄18❤️227👀13103📊93

⚡ Powered by xgo.ing

Today, Fish Audio goes mobile. Introducing Fish Audio iOS, the full voice studio, from desktop to y...

Fish Audio (@FishAudio) — Wed, 26 Aug 2026 17:00:00 GMT

Today, Fish Audio goes mobile. 
Introducing Fish Audio iOS, the full voice studio, from desktop to your pocket.

🐳Discover 2M+ voices 
🔽Create multi-speaker dialogue with 80+ languages 
🔽Direct emotion with open-ended tags 
🔽Clone your own voice, or design one from a single prompt.

Try it now.

Your browser does not support the video tag.

🔗 View on Twitter

💬15🔄21❤️234👀20744📊70

⚡ Powered by xgo.ing

Download here👇 https://t.co/CYfeVpxMx3

Fish Audio (@FishAudio) — Wed, 26 Aug 2026 17:00:00 GMT

Download here👇apps.apple.com/us/app/fish-au…3

💬0🔄0❤️1👀842

⚡ Powered by xgo.ing

https://t.co/LkfV6jWF6i

Fish Audio (@FishAudio) — Tue, 25 Aug 2026 17:30:42 GMT

fish.audio/app/image-vide…

💬0🔄0❤️0👀1153

⚡ Powered by xgo.ing

Try Seedance 2.5 on Fish Creative now!

Fish Audio (@FishAudio) — Tue, 25 Aug 2026 17:30:40 GMT

Try Seedance 2.5 on Fish Creative now!

Fish Creative@FishCreativeHQ

That creative rabbit hole you keep circling? 

Seedance 2.5 just made it a lot deeper. 
Live now on Fish Creative. 

Your idea → 30 seconds of video, with audio.
Jump 🐇

Your browser does not support the video tag.

🔗 View on Twitter

🔗 View Quoted Tweet

💬4🔄6❤️33👀5310📊8

⚡ Powered by xgo.ing

Fish Audio is the official Voice Partner for the Inception II World Model Hackathon 🐳 Build the wo...

Fish Audio (@FishAudio) — Sat, 22 Aug 2026 02:43:48 GMT

Fish Audio is the official Voice Partner for the Inception II World Model Hackathon 🐳

Build the world. We’ll help you give it a voice.

launchd@thelaunchd

@FishAudio is joining Inception 2.0 as our voice partner.

Give your builds a voice :)

🔗 View Quoted Tweet

💬4🔄2❤️24👀3747📊7

⚡ Powered by xgo.ing

#1 most-requested speech model on OpenRouter ahead of @Google . 2M+ builders got us here in 3 weeks...

Fish Audio (@FishAudio) — Fri, 21 Aug 2026 17:04:59 GMT

#1 most-requested speech model on OpenRouter ahead of @Google .

2M+ builders got us here in 3 weeks. And proved the point: the only way to judge a voice is to build with it.

So S2.1 Pro stays free for developers through November.

If you're building anything that speaks, give it a soul.

This is the era of voice, come build with us!

→ fish.audio/blog/s2-1-pro-…

Rissa Cao@rissa_cao

3 weeks ago, we made Fish Audio S2.1 Pro free for developers.

Today it's the #1 most-requested speech model on @OpenRouter ahead of @Google, @grok , @MiniMax_AI  and @Deepgram.

Free wasn't the bad decision. It was our strategy:

🔗 View Quoted Tweet

💬9🔄6❤️87👀6934📊24

⚡ Powered by xgo.ing

👀Spot Fish Audio on your NYC subway commute and tell us which line speaks to you? 🔊

Fish Audio (@FishAudio) — Fri, 21 Aug 2026 15:00:00 GMT

👀Spot Fish Audio on your NYC subway commute and tell us which line speaks to you? 🔊

💬1🔄2❤️20👀2148📊5

⚡ Powered by xgo.ing

We're live on Vercel's AI Gateway.🐟 Free access through Sept 18. Tag us in what you build!

Fish Audio (@FishAudio) — Thu, 20 Aug 2026 04:43:22 GMT

We're live on Vercel's AI Gateway.🐟 Free access through Sept 18. 
Tag us in what you build!

Vercel Developers@vercel_dev

All @FishAudio speech and transcription is free on AI Gateway through September 18th.

𝚐𝚎𝚗𝚎𝚛𝚊𝚝𝚎𝚂𝚙𝚎𝚎𝚌𝚑({
  𝚖𝚘𝚍𝚎𝚕: '𝚏𝚒𝚜𝚑-𝚊𝚞𝚍𝚒𝚘/𝚜𝟸.𝟷-𝚙𝚛𝚘',
  𝚝𝚎𝚡𝚝: '𝚃vercel.com/changelog/fish…�𝚒𝚗𝚐 𝚘𝚞𝚝 𝙰𝙸 𝙶𝚊𝚝𝚎𝚠𝚊𝚢.',
})

https://t.co/KWq0kVI2V3

🔗 View Quoted Tweet

💬4🔄4❤️62👀5537📊13

⚡ Powered by xgo.ing

S2.1 Pro Free is the #1 speech model on @OpenRouter. 1.04M requests last week, up 413% in a week — a...

Fish Audio (@FishAudio) — Wed, 19 Aug 2026 19:00:01 GMT

S2.1 Pro Free is the #1 speech model on @OpenRouter.
1.04M requests last week, up 413% in a week — ahead of every model from Google, x-AI, Microsoft, MiniMax and Deepgram. 

The model is free. The engineering wasn't: 
Custom FP8 CUDA kernels. Continuous batching. 90%+ GPU utilization...Built on an inference stack we rebuilt from the ground up.

💬13🔄9❤️148👀6909📊46

⚡ Powered by xgo.ing

Try S2.1 Pro free for a limited time: https://t.co/egRMsinj7f

Fish Audio (@FishAudio) — Wed, 19 Aug 2026 19:00:01 GMT

Try S2.1 Pro free for a limited time: fish.audio/app/developers/

💬1🔄0❤️6👀1299📊2

⚡ Powered by xgo.ing

Start creating https://t.co/LkfV6jW7gK

Fish Audio (@FishAudio) — Tue, 18 Aug 2026 17:41:41 GMT

Start creating
 fish.audio/app/image-vide…

💬0🔄0❤️3👀1008📊1

⚡ Powered by xgo.ing

We started with voice. Now we’re giving it a whole world to live in🦄 Meet @FishCreativeHQ. The plat...

Fish Audio (@FishAudio) — Tue, 18 Aug 2026 17:14:13 GMT

We started with voice. Now we’re giving it a whole world to live in🦄
Meet @FishCreativeHQ. The platform for creators to turn any ideas into characters, scenes, and ready-to-ship videos. 
Voice, image, video, and lip sync. All in one creative space.

Fish Creative@FishCreativeHQ

Every model can render a face. Making it speak like a person is still the hard part.

Introducing Fish Creative from the audio layer up — the multimodal creative platform by @FishAudio.

→ Voice: generate with S2.1 Pro and inline emotion tags, or clone your own
→ Lip sync: match any face to your chosen voice
→ Image + video: build scenes and motion with leading models, all in one canvas
→ Sound: add any effects on demand

Start with a voice. Build the whole story around it▶️

Your browser does not support the video tag.

🔗 View on Twitter

🔗 View Quoted Tweet

💬11🔄9❤️132👀13739📊33

⚡ Powered by xgo.ing

Try Fish Creative now! 🐳

Fish Audio (@FishAudio) — Tue, 18 Aug 2026 17:05:36 GMT

Try Fish Creative now! 🐳

Fish Creative@FishCreativeHQ

Every model can render a face. Making it speak like a person is still the hard part.

Introducing Fish Creative from the audio layer up — the multimodal creative platform by @FishAudio.

→ Voice: generate with S2.1 Pro and inline emotion tags, or clone your own
→ Lip sync: match any face to your chosen voice
→ Image + video: build scenes and motion with leading models, all in one canvas
→ Sound: add any effects on demand

Start with a voice. Build the whole story around it▶️

Your browser does not support the video tag.

🔗 View on Twitter

🔗 View Quoted Tweet

💬0🔄0❤️1👀157

⚡ Powered by xgo.ing

Connect Fish Audio to your agent. Setup guides here: https://t.co/Keue0KF0Zm

Fish Audio (@FishAudio) — Mon, 17 Aug 2026 21:00:00 GMT

Connect Fish Audio to your agent. Setup guides here: 
docs.fish.audio/overview/mcp

💬0🔄0❤️10👀1791📊4

⚡ Powered by xgo.ing

Meet Fish Audio MCP. Your voices, inside whatever agent you already build in 🪄 — Claude, Codex CLI,...

Fish Audio (@FishAudio) — Mon, 17 Aug 2026 21:00:00 GMT

Meet Fish Audio MCP.
Your voices, inside whatever agent you already build in 🪄
— Claude, Codex CLI, Cursor, and Windsurf.
One endpoint across the agents you already use.

✅Search best voices
✅Generate speech
✅Transcribe audio
✅Check usage

Right inside your workflow. Try it 🎣

Your browser does not support the video tag.

🔗 View on Twitter

💬16🔄26❤️396👀39733📊131

⚡ Powered by xgo.ing

A lot has changed in a year, but we’re still early. 🐳🙌 Grateful for everyone who’s been part of t...

Fish Audio (@FishAudio) — Mon, 17 Aug 2026 19:01:15 GMT

A lot has changed in a year, but we’re still early. 🐳🙌

Grateful for everyone who’s been part of the journey and excited to keep building.

We’re growing the team across GTM and enterprise sales 39-ai.com/careersN2

Rissa Cao@rissa_cao

In just 1 year, Fish Audio reached $21M ARR, 8M users, and raised a $52M seed.

When we started out every investor and friend told me it was too late to compete in voice AI.

Turns out, it wasn't. Here's the story of our first year.

🔗 View Quoted Tweet

💬5🔄1❤️50👀7412📊11

⚡ Powered by xgo.ing

Fish Audio is live in Japan! 🇯🇵 Already heard us in the back of a Tokyo taxi? That’s S2.1 Pro. Ja...

Fish Audio (@FishAudio) — Fri, 14 Aug 2026 15:29:38 GMT

Fish Audio is live in Japan! 🇯🇵 Already heard us in the back of a Tokyo taxi? That’s S2.1 Pro.

Japanese is one of the hardest tests in voice AI. You can pronounce every word correctly and still sound wrong: pitch accent, prosody, keigo, jp ↔ en switching mid-sentence.

→ voice agents for support, sales, scheduling at sub-150ms
→ AI companions with actual personalities
→ game characters: cloning, dynamic dialogue, localization
→ dubbing / animation / audiobooks

8M+ users. 80+ languages. Cloud API or on-prem.
Build with S2.1 Pro free, limited time 👇

Your browser does not support the video tag.

🔗 View on Twitter

💬13🔄8❤️73👀5720📊27

⚡ Powered by xgo.ing

Try S2.1 Pro API for free → https://t.co/QHxmzX2lmw

Fish Audio (@FishAudio) — Fri, 14 Aug 2026 15:29:38 GMT

Try S2.1 Pro API for free → fish.audio/developers/?ut…

💬0🔄0❤️4👀732📊1

⚡ Powered by xgo.ing

Every serious TTS ships emotion tags like [whispering] [laugh] [excited]. Almost nobody uses them - ...

Fish Audio (@FishAudio) — Fri, 14 Aug 2026 01:58:54 GMT

Every serious TTS ships emotion tags like [whispering] [laugh] [excited]. Almost nobody uses them - nobody wants to prompt-wrangle prosody.

Expressive mode fixes that: the LLM writes emotion inline, Fish Audio S2.1 Pro renders it, one toggle in @livekit Agents.

Congrats to the LiveKit team on the launch! And proud of the collaboration that made this sing. 🐟

LiveKit@livekit

1/ Today we're launching Expressive mode in LiveKit Agents.

Most voice agents deliver every sentence in that same, upbeat register we all know and hate (including bad news). Expressive mode makes your agent sound like it registers what the caller is actually feeling.

Your browser does not support the video tag.

🔗 View on Twitter

🔗 View Quoted Tweet

💬7🔄9❤️128👀14559📊35