Haseeb ArshadHaseeb Arshad
← All projects
Speech systemsProduction

Jarvis Voice and Private Transcription

A private speech platform for browser-based voice interaction, uploaded-media transcription, and reusable speech-to-text services.

faster-whisperWeb AudioPythonKubernetesOpen WebUICaddy

1

shared STT service

CPU

portable inference

Private

audio processing

The challenge

What the system needed to solve

Make voice and transcription convenient without uploading recordings or transcripts to a third-party speech service.

The build

Engineering decisions

01

Deployed a reusable faster-whisper API with an ephemeral, re-downloadable model cache rather than treating model files as precious state.

02

Built a standalone transcription lab for file uploads and a voice HUD for conversational interaction.

03

Integrated the same service with Open WebUI and the local agent workstation.

04

Protected browser surfaces with two-factor authentication and constrained the internal API with network policy.

Architecture

How the pieces connect

  1. 1Browser microphone or uploaded media
  2. 2Authenticated voice/transcription UI
  3. 3Shared faster-whisper API
  4. 4Local transcript consumer

Results

What changed

  • Audio and transcripts stay on infrastructure under the operator's control.
  • One speech service supports multiple interfaces without duplicating heavy model state.
  • The model cache can be recreated after failure while user-provided files remain ephemeral.