backchannel
Windows x64
For Windows 10 or later on 64-bit Intel or AMD processors.
- Download
Backchannel-windows-x64-v0.3.0.zip. - Unzip it and run
Backchannel.exe. - If SmartScreen appears, choose More info, then Run anyway.
Release v0.3.0
Voice enrollment, briefing control, and per-call token visibility
Backchannel v0.3.0 adds local voice enrollment for reliable mic-only speaker identity, splits End Call so briefings are generated only when you want them, and shows exactly what each call spent in a post-call Tokens breakdown.
Download Backchannel v0.3.0
These unsigned open-source builds are delivered through the authenticated Backchannel portal. An approved Backchannel account is required; no GitHub identity or repository membership is required.
Voice calibration requires ffmpeg on PATH for desktop bundles.
macOS Apple Silicon
For Apple Silicon Macs. Intel macOS builds are not included in this release.
- Download
Backchannel-macos-arm64-v0.3.0.zip. - Unzip it and move
Backchannel.appwherever you keep apps. - Right-click
Backchannel.appand choose Open the first time.
Linux x64
A portable bundle for 64-bit Intel or AMD Linux systems.
- Download
Backchannel-linux-x64-v0.3.0.tar.gz. - Extract it with
tar -xzf Backchannel-linux-x64-v0.3.0.tar.gz. - Run
Backchannel/Backchannel.
What changed
v0.3.0 focuses on speaker identity you can trust, control over when analysis runs, and transparency into what each call spends.
Local voice enrollment
Enroll your voice once for reliable mic-only speaker identity. Only an encrypted voice embedding is stored; the calibration audio itself is never saved.
End Call, on your terms
End Call is now a split control with End without briefing, disconnects finalize the call with zero analysis, and a briefing can be generated on demand afterward.
Per-call token usage
Every call now tracks its own token usage, summarized in a post-call Tokens breakdown so you can see what each call cost.
Credential checks before the call
Calls refuse to start with an actionable message when the transcription model has no usable credential, instead of failing quietly mid-call.
Split-track retranscription
Retranscription now replays the mic and system tracks separately, preserving who-said-what across segments.
Readable Speakers tab
The post-call Speakers tab is readable and accessible, with honest revalidation of each speaker identity.
Admin About tab
A new Admin About tab shows the running version and release notes, a first-run setup checklist, and a what's-new notice after upgrades.
First-run notes
| Need | What to do |
|---|---|
| Release access | Use the email and password for your approved Backchannel recipient account. |
| Provider key | Add a Gemini key in Admin -> API Keys after launch. Add OpenAI only if you route agents to OpenAI. Calls will not start until the transcription model has a usable credential. |
| Voice calibration | Install ffmpeg and make sure it is on PATH before enrolling your voice; the desktop bundles do not include it. |
| Upgrading | Quit the old version from the tray before starting the new one. Session data is stored per-user and shared by every version. The Admin About tab shows a what's-new notice after upgrading. |
| Desktop launch | Backchannel opens a dedicated local app window when a supported Chromium browser is installed, otherwise it uses the default browser. It remains available from the system tray or menu bar. |
| Docker | Check out the v0.3.0 source tag and rebuild Compose to run the same release through containers. |
Start with the desktop app or self-host the stack
Both paths run Backchannel on your machine. Choose the desktop bundle for the fastest start, or Compose for maximum control.