skald Releases

A feature-rich update adding new routing capabilities like URL generation, pagination middleware, and a new BindError type, along with performance optimizations.

Apr 25, 2026

Apr 9, 2026

This release introduces Spaces, new AI models (Anthropic/OpenAI), and an onboarding email workflow, alongside significant infrastructure migrations to NX and an Effect-based workflow system. It also addresses numerous security vulnerabilities and UI/UX issues.

Apr 8, 2026

This patch release addresses a critical macOS issue where Space and other shortcuts were not detected. It includes fixes for a sidebar shortcut conflict, support for virtual microphones, and security updates for dependencies.

Apr 4, 2026

This major release introduces Voice Mode for hands-free control, remote LLM server support, and full settings management via CLI. It also expands shortcut capabilities and improves the overall user experience.

Mar 29, 2026

Murmure 1.7.0 pushes the application into the LLM-powered workflow era with major text manipulation features and a strong focus on privacy by design. It introduces a Select Command for contextual editing, improved security through SonarQube scanning, and officially signs macOS and Windows binaries.

Feb 12, 2026

This release introduces local LLM connectivity via Ollama and advanced text formatting rules, transforming Murmure into a more versatile tool. It also adds a Toggle-to-Talk mode, sound feedback, and experimental support for Intel-based macOS.

Dec 15, 2025

This release introduces a new parameter for detecting document rotation, enhancing the parsing capabilities for files that may be skewed or rotated.

Aug 29, 2025

This update focuses on performance and stability, featuring reduced retry logic, fixes for color space handling in PDF images, and the addition of metadata to parsed documents.

Aug 22, 2025

This version introduces an early preview of confidence scores for field extraction, allowing users to gauge the reliability of extracted data, alongside various test updates.

Jul 21, 2025

A maintenance release that addresses critical bugs related to file encoding on Windows and image cropping boundaries, while adding a new configuration object for API settings.

Jul 9, 2025

This release focuses on optimizing the caching layer, introducing an asynchronous resolver workflow, and standardizing the charset encoding system with new presets.

May 21, 2025

This update expands document support, introduces asynchronous runtime balancing, and adds export/import capabilities for worker indexes and documents.

Mar 24, 2025

A major version release focusing on persistent storage support across multiple databases, significant performance improvements, and the introduction of a robust new resolver system.

Mar 17, 2025

Introduces a new graphical user interface (GUI) that allows users to precisely control utterances by generating, modifying, and refining realizations across over 7000 languages.

Oct 7, 2024

This minor release significantly improves audio quality by retraining the vocoder end-to-end on synthetic spectrograms and increasing its capacity while maintaining fast inference speed.

Sep 22, 2024

Enhances the 7000-language Text-to-Speech model with stochastic prosody prediction, expanded IPA modifier support, additional pretraining languages, and an overhaul of language similarity modules.

Jul 25, 2024

A major milestone release enabling Text-to-Speech (TTS) in over 7000 languages by extrapolating a pretrained checkpoint, adding audio watermarking, and improving overall quality.

Jun 10, 2024

Adds emotional control to the TTS model via prompting, allowing users to condition the model on emotional prompts during training and transfer emotions to synthesized speech during inference.

Jun 10, 2024

Oct 3, 2022