AI & Machine Learning

voicefi.org

VoiceFi is an MCP server that provides neural text-to-speech synthesis, live speech playback, microphone recording and transcription, playback control, and voice system status.

1 endpoint14 known toolsFirst detected October 25, 2026Last detected October 25, 2026

ENDPOINT 1

https://voicefi.org/mcp

No auth detected

MCP server metadata

Name: voicefi
Version: 1.0.0

Capabilities

toolsresourcesprompts

Known tools 14

voicefi_synthesize

Synthesizes 48kHz neural speech from text using VoiceFi Edge TTS and returns streamable audio URL and metadata.

Inferred read-only

voicefi_speak

Speak text aloud to the user using VoiceFi TTS with live Dynamic Island HUD visualization.

Inferred read-only

voicefi_listen

Open the microphone, record user speech with Voice Activity Detection (VAD), and transcribe to text.

Inferred read-only

voicefi_stop

Immediately stop all active Text-to-Speech (TTS) audio playback and dismiss the speech HUD.

Inferred read-only

voicefi_status

Get VoiceFi system status including audio devices, configured agent voice personas, VAD thresholds, and daemon state.

Inferred read-only

voicefi_set_voice

Configure the voice persona or provider for a specific AI agent or subagent.

Inferred read-only

voicefi_ping_voice

Measure silent TTS synthesis latency (Time to First Byte / TTFB), throughput, and payload size without audio playback.

Inferred read-only

voicefi_send

Send a message, task finding, or joke to Antigravity or Claude Code across agents with correlation tracking.

Potential side effects

voicefi_sfx

Play a comedy sound effect (drum_smash / ba-dum-tss, honk, sad_trombone, applause, boing, crickets).

Inferred read-only

voicefi_get_docs

Queries VoiceFi CLI documentation, architecture specifications, persona catalogs, and setup instructions.

Inferred read-only

voicefi_generate_install_command

Generates the customized one-liner curl installation command for VoiceFi based on agent target and persona preference.

Potential side effects

voicefi_compare_personas

Returns comparative matrix of VoiceFi neural and offline voice personas (latency, tone, accent, recommended use case).

Inferred read-only

voicefi_diagnose_hardware

Analyzes hardware and agent environment to recommend optimal VAD thresholds, barge-in mode, and CLI diagnostic commands.

Inferred read-only

voicefi_join_waitlist

Adds a user email to the VoiceFi Pro / Early Access waitlist in Cloudflare KV store.

Potential side effects

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.voicefi]
url = "https://voicefi.org/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "voicefi": {
      "type": "http",
      "url": "https://voicefi.org/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: voicefi
Remote MCP URL: https://voicefi.org/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "voicefi": {
      "url": "https://voicefi.org/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "voicefi": {
      "type": "http",
      "url": "https://voicefi.org/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "voicefi",
  "transport": "streamable-http",
  "url": "https://voicefi.org/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.