DEVMAN

Voice-Only Coding

Talk your code into existence. Devman is an AI voice assistant that takes your coding tasks and delegates them to your dev environment — entirely through natural conversation.

Stored using JJFS

Setup lives local

DEVMAN

Voice-Only Coding

ALPHA

🔗

Connect data2link

Enter your data2link API key. Devman uses it to store your settings and power AI. You can create a key at data2l.ink.

data2link API Key

🔑

One more thing

Devman uses ElevenLabs for voice. Paste your API key below to enable speech synthesis and recognition. You can update this anytime in settings.

ElevenLabs API Key

Say "Hey Devman", tap the button, or type below

Scroll to bottom

Settings

Voice Settings

ElevenLabs API Key

Voice ID

Voice Model

Wake word in-browser (w/o Wand Awake)

Hiatus mode (ignore Wand Awake)

Dead Air Song

Beep-Boop Volume

60%

Enrolled Speakers

No speakers enrolled yet. Say "Hey Devman, adapt to my voice" to enroll.

Speaker Recognition Thresholds

Recognition Threshold (for "Hey Devman")

55%

Higher = stricter matching. Default: 55%

Enrollment Threshold (auto-match)

30%

Lower = more forgiving. Default: 30%

Connected Apps

No apps registered.

Add Devman-Compatible App

Account

data2link API Key

Advanced

Auto-stop after of no input

Verbosity

Text Size

Agent mode (enables tool use, slower)

Timeout when WebSocket stop signal received

Dev Environment

Dev Environment Endpoint (session protocol)

Dev Key (auth for your dev environment — sent as Bearer)

Project ID (sent as project_id when opening a session)

Shared with the Devman home page — stored in this browser as devman_settings. See Docs for the protocols.

D2L Debug

0 requests