Brieflee speech and speaker model files: licence notice Published 6 October 2026 at https://brieflee.ai/models/LICENSE.txt The files under https://brieflee.ai/models/ are downloaded by the Brieflee phone app and the Brieflee desktop app and run on your own device. They are of two kinds, and the two kinds carry different terms. 1. Brieflee's own models The files whose names carry "brieflee" (for example ggml-small-brieflee-6-q8_0.bin and ggml-base-brieflee-2-q8_0.bin) are Brieflee's fine-tuned speech models. They were built from OpenAI's Whisper weights and trained further on publicly available recordings only, never on a customer's meeting. Copyright (c) Xtreme IT Services, trading as Brieflee. They are licensed to you, not sold, for use with your Brieflee account, the Brieflee apps and the Brieflee service, under the Terms of Service at https://brieflee.ai/terms. You may not copy them elsewhere, redistribute them, use them in another product, or extract, de-quantise or train from them. The MIT notice for the Whisper weights they are built from is reproduced below and continues to apply to that part. 2. Open-source models served under their own licences These files keep the licence their authors gave them. The Brieflee terms in part 1 do not apply to them and nothing in this notice narrows the rights those licences give you. - ggml-tiny-q5_1.bin, ggml-base-q5_1.bin, ggml-small-q5_1.bin, ggml-small-q8_0.bin: OpenAI's Whisper weights (MIT licence, Copyright (c) 2022 OpenAI), converted to the ggml file format with whisper.cpp (MIT licence, Copyright (c) The ggml authors). - pyannote-segmentation-3-0.onnx: pyannote speaker segmentation 3.0 (MIT licence, Copyright (c) CNRS), as released in ONNX form by the sherpa-onnx project. - wespeaker-resnet34-lm-gm.onnx: the WeSpeaker ResNet34-LM speaker embedding model, trained on VoxCeleb by the WeSpeaker project and released under the Creative Commons Attribution 4.0 International licence (https://creativecommons.org/licenses/by/4.0/), as released in ONNX form by the sherpa-onnx project. Brieflee added one metadata entry (feature_normalize_type) and changed nothing else. The MIT licence (OpenAI Whisper, whisper.cpp, pyannote segmentation) Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. Questions about this notice: legal@brieflee.ai