The structure and portions of the local speech-to-text implementation in stt/ were adapted from
OpenWorker's ocw-stt crate.
Copyright (c) 2024 Andrew Ng
Licensed under the MIT License. The full license text is available in the OpenWorker source distribution and at https://opensource.org/license/mit.
Streaming speech recognition, voice activity detection, and text-to-speech use the sherpa-onnx Rust API and native runtime.
Copyright (c) the sherpa-onnx contributors.
Licensed under the Apache License 2.0. See https://github.com/k2-fsa/sherpa-onnx.
Downloaded speech models remain under the licenses distributed inside their model archives.
Acoustic echo cancellation, noise suppression, high-pass filtering, and automatic gain control use Sonora's pure-Rust port of the WebRTC Audio Processing Module.
Copyright (c) the Sonora and WebRTC contributors.
Licensed under the BSD 3-Clause License. See https://github.com/dignifiedquire/sonora.
The pi-server bundles pptxgenjs, used by agents to generate native PowerPoint files.
Copyright 2015-2022 Brent Ely and pptxgenjs contributors.
Licensed under the MIT License. See https://github.com/gitbrent/PptxGenJS.
PPTX layout-issue inspection, OOXML validation, and high-fidelity preview rendering use the
OfficeCLI binary, distributed via the @officecli/officecli npm package.
OfficeCLI, Copyright 2026 OfficeCLI (https://OfficeCLI.AI), created and maintained by goworm.
Licensed under the Apache License 2.0. Per its NOTICE, redistribution with or without modification must retain this notice. See https://github.com/iOfficeAI/OfficeCLI.