mortiphi

The description is self-reported and unverified. No revenue, customers, traction or independent validation are evidenced. The project appears to be a hackathon submission with no demonstrated commercial traction.

The description states mortiphi:

• Turns any OpenCode thread into a voice agent with one command
• Uses the command /mortic to start speaking in an ephemeral fork of the main thread
• Populates a sidepod in the TUI and starts a popup in the GUI
• Inherits the main thread's reasoning
• Uses full duplex architecture with acoustic echo cancellation and eager end of turn
• Handles near-instant self-compaction using Mercury 2 (a diffusion model)
• Has a speak-screen bipartite scheme in structured output
• Works only on OpenCode for now

The author describes it as a conversational interface that fits how normal human conversation would be like.

The description states:

• The product was inspired by the need to understand agent thinking through voice communication instead of text
• It evolved from side-chats in Codex to a voice-native coding interface
• The author's original approach used Codex App Server but pivoted due to latency issues
• They built it for OpenCode using GPT 5.6 Sol, primarily to refine the harness and nature of speak-screen response

The positioning appears to be: a voice interface for coding agents that improves upon text-based side-chats by enabling natural conversation flow.

Not evidenced. The description does not state who the target customer is beyond "users" or "developers". No specific customer segments, personas or buyer profiles are mentioned.

Not evidenced. The description does not contain any information about pricing, revenue model, monetization strategy or business model.

The description states:

• Built with bun, cartesia, deepgram, fastapi, inception-labs, javascript, livekit, macos, mercury-2, node.js, opencode, opentui, portaudio, python, rest-api, server-sent-events, solidjs, sqlite, typescript, uvicorn, voice, webrtc, websockets, zod
• Uses GPT 5.6 Sol for OpenCode
• Implements full duplex architecture with acoustic echo cancellation and eager end of turn
• Uses Mercury 2 (a diffusion model) for self-compaction
• Has a speak-screen bipartite scheme in structured output
• Handles markdown suppression and code vs speak handling
• Processes complex prompts within 2 seconds

The description states:

• Usage costs for Mercury 2 + STT + TTS are lower than GPT Realtime 2.1 API, offering nearly triple the number of hours
• They now use Mortic for their own use cases
• Can process complex prompts within 2 seconds
• Have almost never run into speak-screen issues during normal usage
• Pivoted entirely from original Codex App Server approach

However, no evidence of external users, customers or adoption beyond the author's own use cases is provided.

Not evidenced. The description does not mention any competitors or competitive landscape.

• The project appears to be a hackathon submission with no demonstrated commercial traction
• No evidence of revenue, customers or adoption beyond the author's own usage
• The product is limited to OpenCode only
• The author states they had to rebuild the entire agent and harness due to previous version being unreliable
• The description is self-reported and unverified

1. What is your actual user base beyond yourself?
2. How many users are currently using this in production?
3. What is your monetization strategy?
4. How do you plan to expand beyond OpenCode?
5. What are the technical limitations of this approach that prevent broader adoption?
6. How does this compare to existing voice interfaces for coding tools?
7. What are your plans for scaling beyond the current hackathon prototype?