Two distinct speakers
Assign a preset voice to each side of the conversation.
MOSS TTSD
MOSS TTSD: AI Dialogue & Podcast Voice Generator
Generate multi-speaker audio with MOSS TTSD. Write a dialogue, assign two voices and create podcast scenes, interviews or conversational lessons.
6 credits / 1,000 charactersYour credits work across all models.
Your output
A little input. A new possibility.
Your finished audio, image or text will appear here.
01 / Inside the model
More than one voice. One flowing story.
MOSS TTSD is built for multi-speaker speech rather than a single narrator. The official research model explores long conversations, speaker continuity and multilingual dialogue. Our workspace provides a focused two-speaker workflow: write labelled turns, choose one preset voice for each speaker and generate the conversation as audio. Shape the exchange in the script, then listen for pacing, pronunciation and clear turn-taking.
Scripted turn-taking
Use speaker labels to control who says each line.
One finished recording
Generate and download a conversation as a single audio result.
02 / Examples & listening room
Explore the official modelSee it. Hear it. Get a feel for it.
Official examples are credited to OpenMOSS. Studio samples were generated by us with the named API model. These are saved previews, not a result of your current settings. Research capabilities and available studio controls may differ.
A question worth asking
Studio sample · English03 / Use cases
Where it fits.
Podcast drafts
Test an interview, a co-host introduction or a scripted discussion.
Conversation lessons
Create a two-person role-play for learning and rehearsal.
04 / Questions
A few things worth knowing.
How do I format a dialogue?
Use the example button for the required speaker-labelled format, then replace the lines with your own script.
How are dialogue credits calculated?
6 credits per 1,000 spoken characters. Speaker labels are excluded from the billed spoken text; the total rounds up to a whole credit.
Can I use five speakers here?
Not yet. The official research model describes that capability; this workspace offers two preset speakers.
Made for your next idea
Choose a model, add your input and review the credit cost. Successful results stay in your account history; AI output may be inaccurate.
Text is rounded up to a whole credit. Audio is billed per started minute. Live voice uses one credit per 6 seconds.
Failed tasks are refunded automatically.
Result links can expire. Download files you want to keep.
Model documentation