Soliloquy
Voice-to-voice AI roleplay
5 followers
Voice-to-voice AI roleplay
5 followers
Soliloquy is a voice-to-voice AI cockpit for practicing spoken scenes out loud. Forge a character, hit Speak, and you're in a real conversation — emotion, interruption, accent, all live. Built for voice actors, dungeon masters, and sales-training teams.
Interactive
Free Options
Launch Team / Built With
Render Sandboxes: a computer for every agentGive every agent its own machine. Starts in under a second
Promoted
Hey Product Hunt 👋
I built Soliloquy because every "AI roleplay" tool I tried was just typing in a chat box. Voice actors don't audition by typing. Dungeon masters don't run NPCs by typing. Sales reps don't pitch by typing. So I made the thing that's missing: you press a button, you talk, the character talks back — in real time, in character, in their voice.
What you can do today
🎭 Forge a character — name, archetype, objective, backstory, vocal style. Or pick from 8 starter personas (noir detective, furious enterprise customer, elven black-market merchant, etc.)
🎙️ Speak into your mic — Whisper-1 transcribes, Claude Sonnet 4.5 stays in character with persona-aware replies, 9 synthetic voices speak back. Turn latency ~2.1s.
✂️ Interrupt them mid-sentence — the audio stream stops the moment you start talking again, like a real scene partner
🎚️ Mix the voice — pitch, reverb, EQ, compressor, de-esser, distortion, telephone filter, stereo width, echo. All client-side. Studio tier gets the full DAW.
🎞️ Save Reels — clip the best lines from a session into a downloadable MP3. Useful for demo reels, session highlights, training playback.
🌍 Multi-language — every voice speaks every language. Switch tongue mid-scene.
🆓 Try without signup — 3 free turns at trysoliloquy.com/try
Who it's for
Voice actors practicing difficult character transitions when no scene partner is around
Dungeon masters / TTRPG GMs automating NPC interaction so the table flows
Corporate training / sales enablement drilling reps against a stubborn customer persona at 2am
The unfair edge: nobody else is doing real-time voice-to-voice with persona drift correction. Most "AI character" tools type back. The few that speak feel like reading a teleprompter. Soliloquy commits to the character — including mid-scene tone shifts, interruptions, and emotional escalation.
Free tier gives you 30 turns/month + the starter library. Pro/Studio unlock ElevenLabs voices, NSFW personas, marketplace publishing, Reels export, and the full DAW.
I'll be in the comments all day. Throw the weirdest persona idea you can think of at me — I'll fork it and post the audio in the thread.
— Maker of Soliloquy 🔗 trysoliloquy.com