Resonant 0.5.0

Resonant 0.5.0

Resonant 0.5.0 turns local AI music generation and personal singing-voice conversion into first-class workstation capabilities.

Highlights

• Personal singing voices with Seed-VC: install and manage the optional local provider from Resonant, select an isolated source vocal and a clean 1–30 second reference, and add the converted performance directly to the Audio track.
• Consent and provenance by design: every conversion requires confirmation that the voice is owned or used with explicit permission. Resonant records source/reference hashes and conversion provenance beside the output.
• Visible provider controls: the main header now exposes labeled AI SONG and MY VOICE controls with install, ready, running, and error states instead of hiding providers behind icon-only actions.
• Managed local setup: Resonant detects existing managed ACE-Step and Seed-VC installations, offers installation when missing, and coordinates GPU use by stopping ACE-Step before Seed-VC conversion.
• Agent-accessible voice conversion: MCP clients can inspect Seed-VC availability and perform consent-gated conversions using workspace-confined WAV paths.

Seed-VC integration

• The managed Windows installation pins archived Seed-VC revision 51383efd921027683c89e5348211d93ff12ac2a8 and installs it separately under its GPL-3.0 license.
• Windows setup installs explicit PyTorch 2.4 CUDA 12.4 wheels and a minimal singing-inference dependency profile, avoiding malformed upstream requirement flags and unnecessary GUI/evaluation packages. Partial installations remain retryable and are not reported as complete.
• Installer progress strips terminal control codes, preserves a readable failure message, and removes temporary package caches after successful setup.
• The singing/F0 conversion path retains source melody, timing, phrasing, and pitch contour while transferring vocal identity from the reference recording.
• Source, reference, and output files are restricted to the configured workspace for MCP operations.
• An NVIDIA CUDA GPU with 6 GB VRAM is the practical minimum; 8 GB is recommended. Model weights download on first conversion and remain local afterward.

MCP additions

• get_voice_conversion_provider_status reports installation, runtime, hardware, and readiness state.
• start_voice_conversion, get_voice_conversion_status, collect_voice_conversion, and abandon_voice_conversion provide a durable job lifecycle that survives request timeouts and MCP reconnects. Stable idempotency keys prevent duplicate GPU work.
• convert_singing_voice remains available as a blocking compatibility tool.
• Conversion controls include diffusion quality, pitch shift, guidance strength, and 0.5×–2× performance-length adjustment.
• Capability metadata explains the voice-conversion workflow, hardware expectations, provider license, consent boundary, and isolated-vocal requirement.
• Newly generated Codex MCP configuration allows ten minutes per tool call so full-song local vocal analysis can complete on CPU; long AI generation and voice conversion still use the durable start/poll/collect workflows.

Validation

• Delivery-graph validation, TypeScript checks, and the complete unit/domain suite pass.
• Production renderer and MCP builds pass.
• MCP smoke validates all registered tools, including Seed-VC provider status and consent-gated conversion contracts.
• Electron smoke covers provider discovery, visible header states, the Seed-VC installation experience, responsive layouts, and the existing workstation workflows.

Downloads

• Resonant-Setup-0.5.0-Windows-x64.exe — standard Windows installer.
• Resonant-0.5.0-Windows-x64.exe — portable Windows build.
• SHA256SUMS.txt — checksums for verifying both executables.

The Windows executables are currently unsigned, so Windows SmartScreen may show a warning. Source installation instructions remain available in the README.

Resonant 0.4.2

Resonant 0.4.2

Resonant 0.4.2 adds safe audio-conditioned ACE-Step generation for connected MCP agents.

What changed

• MCP clients can pass an existing workspace WAV to ACE-Step as reference audio for global timbre and style guidance, including approximate vocal character.
• Cover generation can use a workspace WAV as structural source audio with adjustable influence strength.
• Repaint generation can use a source WAV plus a bounded time interval and optional edit instruction.
• Audio inputs are resolved through Resonant’s confined workspace boundary. Missing files, non-WAV paths, traversal attempts, and symlink escapes are rejected before ACE-Step starts work.
• Invalid task combinations are rejected early, including covers without source audio, source audio attached to plain text-to-music jobs, and inverted repaint intervals.
• get_capabilities and the MCP tool schemas now explain audio-conditioning modes, required inputs, and the limits of reference-based vocal identity.
• README and agent-production documentation now cover recording/reference workflows and voice-consent expectations.
• Regression coverage verifies ACE-Step payload mapping, confined path resolution, required source inputs, task-specific controls, and repaint interval validation.

Validation

• Delivery-graph validation, TypeScript checks, and the complete 66-test unit/domain suite pass.
• Production renderer and MCP builds pass.
• MCP smoke validates all 29 registered tools, revision safety, workspace confinement, analysis, rendering, and the rebuilt audio-conditioning schema.
• Electron smoke covers startup, instruments, recording/import, arrangement, mixing, save/open/recovery, playback, and WAV export.
• Packaged MCP smoke validates the server bundled inside the Windows application.

Downloads

• Resonant-Setup-0.4.2-Windows-x64.exe — standard Windows installer.
• Resonant-0.4.2-Windows-x64.exe — portable Windows build.
• SHA256SUMS.txt — checksums for verifying both executables.

The Windows executables are currently unsigned, so Windows SmartScreen may show a warning. Source installation instructions remain available in the README.

Resonant 0.4.1

Resonant 0.4.1

Resonant 0.4.1 is a reliability update for AI-assisted songwriting and local ACE-Step generation.

What changed

• Multilingual section labels are normalized for ACE-Step while the original native-language lyrics remain intact in the Resonant project.
• Korean and other installed songwriting packs can now pass their language context through the complete MCP generation workflow.
• ACE-Step generation runs independently from the MCP client process, so a long-running song job continues if Codex, Claude Code, or another client disconnects.
• Interrupted or abandoned generation jobs can be restarted safely without creating duplicate work.
• ACE-Step status is scoped to an installed runtime, so a service running elsewhere no longer makes an empty installation location appear active.
• MCP documentation now includes a simple assistant-guided setup path: share the repository URL with an AI assistant and ask it to configure Resonant.
• Regression coverage was added for multilingual prompt normalization, detached generation, abandoned-job recovery, and language propagation.

Validation

• Delivery-graph validation, TypeScript checks, and the complete unit/domain suite pass.
• Production renderer and MCP builds pass.
• MCP smoke validates all 29 registered tools, songwriting-language discovery, revision safety, confinement, analysis, rendering, and AI-generator status.
• Electron smoke covers startup, instruments, audio, arrangement, mixing, save/open/recovery, playback, and WAV export.
• Packaged MCP smoke validates the server bundled inside the Windows application.
• A live Korean ACE-Step job was started, reconnected after the originating MCP client exited, and completed successfully.

Downloads

• Resonant-Setup-0.4.1-Windows-x64.exe — standard Windows installer.
• Resonant-0.4.1-Windows-x64.exe — portable Windows build.
• SHA256SUMS.txt — checksums for verifying both executables.

The Windows executables are currently unsigned, so Windows SmartScreen may show a warning. Source installation instructions remain available in the README.

Resonant 0.4.0

Resonant 0.4.0

Resonant 0.4 adds an optional local generative-music engine and a substantial visual refresh while keeping ordinary projects and startup lightweight.

Highlights

• Optional ACE-Step 1.5 post-install module with isolated dependencies and shared model storage outside project folders.
• Hardware-aware RTX 4060 Laptop profile: ACE-Step 2B Turbo, 0.6B planner, PyTorch backend, batch size 1, and CPU offload.
• New ACE-Step Studio for prompting instrumental or vocal music, choosing duration and tempo, and adding generated WAVs directly to the Audio track.
• Engine start, stop, storage reporting, generated-output count, and complete independent removal controls.
• MCP get_ai_generator_status and generate_ai_music tools for agent-driven local generation followed by Resonant import, arrangement, mixing, analysis, and rendering.
• Copy-ready MCP onboarding for Codex, Claude Code, generic STDIO clients, and PowerShell, with workspace confinement and sanitized examples.
• Songwriter Studio language packs for English, Standard Mandarin, Hindi, Korean/K-pop, Latin American Spanish, and Japanese, exposed through both the desktop app and MCP discovery.
• Three creative starters: Neon Raga, Midnight Drive, and Velvet Orbit.
• Premium dark workstation refresh with clearer panel depth, teal/violet hierarchy, stronger contrast, and offline-safe Space Grotesk/Inter/JetBrains Mono font stacks.
• New Reso mascot and theme-aware Resonant wordmark for the public project page.
• Improved hover, focus, disabled, scrollbar, modal, transport, clip, mixer, and keyboard-accessibility states.
• Model caches remain centralized so multiple projects never duplicate the multi-gigabyte runtime.

Validation

• Unit/domain suite, TypeScript checks, delivery-graph validation, and production build pass.
• MCP smoke covers all 29 registered tools, language-pack discovery, optional AI-generator status, path confinement, revision conflicts, deterministic render, analysis, and undo.
• Electron smoke covers the optional ACE-Step post-install screen plus instruments, Flow, audio import/record, Arrange, Mix, save/open/recovery, playback, and WAV export.
• Real ACE-Step model installation and generated-sample measurements are recorded with the release artifacts.