Screenshots
Watch trailer on Steam ↗Beyond Voice
Nikolaus Zaoputra · Published by Nikolaus Zaoputra
Content notice: Due to the nature of the product with the conversation response to be AI generated, there is possibility that it might respond in a way that involves mature or sensitive subjects. Please do not use the product if you are not in legal age for mature content in your country, or in sensitive mental condition.
3 user reviews — 67% positive (3 reviews)
Last updated 28 Sept 2026
Beyond Voice is a voice-first digital companion: speak naturally, get voiced replies with lip-sync and subtle gestures, and watch your relationship, mood, and shared memories evolve over time—built for quick sessions or long chats. Your relationship and memories persist across sessions.
£4.29
Lowest price we've ever recorded
View on SteamPrices can lag the live Steam store — always confirm before buying.
Price history
Indie, Simulation, Early Access · 2026 · More under £5 games
Delve into Nikolaus Zaoputra’s innovative Beyond Voice, where your words shape a digital companion’s responses, emotions, and shared history. Experience this unique early access simulation that thrives on voice interaction for both quick chats and deep conversations.
About this game
Beyond Voice is a small, focused chat experience. If you enjoy having discussion with AI, but wants a more visual experience, this app answers that need. All of the features runs completely local on your PC; no worry about per-use subscription fee or uploading your personal conversation to the cloud.
Meet Jenny
She is the AI avatar that will be your personal conversation partner. Not only does she respond to your questions, she also keeps structured memory so conversations maintain continuity across days: recent conversation context, longer-term summaries, and simple personal facts you’ve shared.
Over time, Jenny develops an evolving relationship and mood toward you—becoming warmer, more distant, or more playful based on how your conversations go.
Keep your tone positive and show interest to her suggestion, she will become warmer and more accepting of intimate interactions. Be rude or keep ignoring her, you'll regret that this game goes beyond voice...
Key Features
Talk with your voice: press-to-talk or auto-listen; your speech is transcribed and answered in real time. server
Spoken replies + performance: responses come back as voiced audio, with lip-sync and talk animation support designed for a screen character.
Relationship + mood engine: the companion tracks relationship, valence, and engagement internally and uses them to shape tone and behavior over time.
Memory that persists: recent turns are saved, and longer-term memory is maintained so conversations can reference what happened previously.
Light “conversation beats”: as trust increases, optional personal topics can unlock—short, guided moments that test empathy and accountability (kept subtle; not a traditional quest log).
Runs completely local: It’s a self-contained experience on your PC—no cloud-service grind.
[Feb 16, 2026 Update] Bring-Your-Own-LLM! You can now use your own preferred model to better match your interaction preference! Check the README file for instruction.
Note: To enable quick response time of the conversation in a fully offline setup, most of the critical processes utilizes Nvidia's CUDA libraries. Nvidia GPU with minimum of 16GB VRAM is required to run the game properly.
System Requirements
Will my PC run this? →Minimum
- Requires a 64-bit processor and operating system
- OS: Windows® 10 64-bit
- Processor: Intel Core i5-13400 | AMD Ryzen 5 7600 or higher
- Memory: 16 GB RAM
- Graphics: GeForce RTX 5060 Ti or higher
- DirectX: Version 12
- Storage: 13 GB available space
- Sound Card: DirectSound® supported sound card
- Additional Notes: NVIDIA GPU with minimum of 16GB VRAM is REQUIRED. Use of multiple NVIDIA GPU also works.
Recommended
- Requires a 64-bit processor and operating system
- OS: Windows® 11 64-bit
- Processor: Intel Core i5-13500 | AMD Ryzen 7 7700 or higher
- Memory: 32 GB RAM
- Graphics: GeForce RTX 5060 Ti or higher
- DirectX: Version 12
- Storage: 13 GB available space
- Sound Card: DirectSound® supported sound card
- Additional Notes: NVIDIA GPU with minimum of 16GB VRAM is REQUIRED. Use of multiple NVIDIA GPU also works.
Supported Languages
*languages with full audio support
Latest news
Patch NotesOfficial
March 2nd Update
Update contents: Added avatar greeting when the game initially starts. Significantly improved response time. Speech now starts much earlier even for very long responses. Added slight variation in speaking speed depending on sentence context.Read full article on Community Announcements (Dying Cat)Patch NotesOfficial
Multiple updates: up to Feb 25th 2026
\- Expression change now transitions smoothly instead of abruptly changing.- Added recording progression bar during auto-listen mode to better reflect length limit.- Improved character prompt to make conversation more natural.- Improved memory logic to keep more important facts in long term memory.\- Fixed issue with relationship level mechanism to unlock higher relationship level.- Relationship level cap changed to just 1 stage, capped only at 8.0.- Added expression change to avatar's face depending on mood level. Avatar's response is now better tuned to its status.- Improved interaction scoring.- Enabled KV cache on LLM backend to improve response time.- Improved communication protocol between backend processes for stability.\- Improved variation in avatar reaction depending on relationship level- *Known issue* Relationship level cap unlock mechanism not working, currently not possible to go past score of 7.0.\- Added option to use custom LLM (limited to GGUF)- Fixed some animation transition issue- Added README documentRead full article on Community Announcements (Dying Cat)
