All notable changes to VocalizeAI are documented in this file.
Format follows Keep a Changelog 1.1.0. Versions follow Semantic Versioning.
No unreleased changes yet.
1.0.0 — 2026-05-18
First public release. Apache 2.0 open-source milestone.
Internal source: d9bd923 · Public repo first commit: 591aa9e
Release notes: https://github.com/DGPisces/VocalizeAI/releases/tag/v1.0.0
- Universal phone-task engine — 5-layer dialogue pipeline (preflight, question, translate, summarize, merchant) handles arbitrary phone-call scenarios without per-task configuration.
- Bilingual zh/en UI — React frontend with full Chinese and English locale support from day 1.
- Raspberry Pi orchestrator — production-grade deploy target; install script
- systemd unit + cloudflared tunnel for zero-port-forward remote access.
- Audio loopback testing — Playwright + pytest harness drives laptop-loopback calls and post-call callback flows; 8 text-bypass AI merchant scenarios with a deterministic judge.
- Apache 2.0 license — OSS launch with SECURITY.md, CONTRIBUTING.md, issue templates, PR template, and CODEOWNERS.