Get 50% off for your first 30 days. Drama 3 is Fish Audio's most emotive, controllable model yet. Write a multi-character scene as one conversation, then build it with audio in a single workflow.
Step 01
Open Comfy Cloud
Sign in or create a free account
Step 02
Select Drama 3
Add the Fish Audio Text to Speech node from partner nodes and set model to drama-3
Step 03
Direct the scene
Add your voices, write the dialogue, and run it
Fish Audio's newest model, and the one it calls its most emotive and controllable yet. You write a multi-character scene and set the direction, and Drama 3 performs it as a single conversation. It runs in the Fish Audio Text to Speech node, in Comfy Cloud and in local ComfyUI 0.39.3 or newer.
Open Comfy Cloud and search the Template Library for Fish Audio Drama 3, or add the Fish Audio Text to Speech node from partner nodes and set model to drama-3. Add your voices, write the dialogue, and run it.
Drama 3 is 50% off for your first 30 days in Comfy Cloud. Open Comfy Cloud to start.
Drama 3 runs in the Fish Audio Text to Speech node, so it fits into a Comfy workflow like any other node. Build the scene and its audio in one graph. Change a line, run again, and everything downstream picks up the new take.
No. Drama 3 runs through Fish Audio's API, so there is nothing to download and no GPU to provision.
Text to Speech also runs S2.1 Pro and S1, which take emotion cues in the text. Speech to Text transcribes audio with automatic language detection, and Instant Voice Clone makes a private voice from 1 to 20 recordings. Voice Selector is where you pick a voice from the Fish Audio library, and all four are in the Fish Audio docs.
Yes, with the Fish Audio Instant Voice Clone node: connect 1 to 20 recordings, under 270 seconds in total and ideally 10 to 30 seconds each, and the cloned voice feeds straight into the Fish Audio Text to Speech node with no training wait. Voice cloning is free. Clone only voices you have permission to use.
Per use, through your ComfyUI account. Text to speech is metered by text length and speech to text by the minute. Current rates are in the Fish Audio docs.
Access cloud-powered ComfyUI workflows with straightforward, usage-based pricing.
Launch offer: 50% off Fish Audio Drama 3 for your first 30 days
5 free runs on real GPUs — no credit card required.
Billed monthly
What's included:
Comfy Agent
Comfy API
Generates ~380 5s videos*
Billed monthly
Everything in Standard, plus:
Comfy Agent
Comfy API
Generates ~670 5s videos*
Billed monthly
Everything in Creator, plus:
Comfy Agent
Comfy API
Generates ~1,915 5s videos*
Built for teams collaborating on workflows together.
Billed monthly
Generates ~13,405 5s videos*
Everything in Pro, plus:
Comfy Agent
Comfy API
Coming soon...
Need more members? Looking for more flexibility or custom features?
*Based on 5s videos created with the Wan 2.2 Image-to-Video template using default settings (81 frames, 18fps, 640x640, 4-step sampler)
Your AI assistant can access the ecosystem, build workflows, and generate images, video, audio, or 3D.
Run Drama 3 in Comfy Cloud today, batch scenes through the API, or run it from ComfyUI on your own machine.