Evoo
Copyright 2026 Evoo contributors
Licensed under the Apache License, Version 2.0.

Evoo downloads and runs the following third-party open models and libraries.
Model weights are downloaded on first use and are not distributed with Evoo.

Speech recognition
- NVIDIA Parakeet TDT 0.6B v3 — CC-BY-4.0 — https://huggingface.co/nvidia/parakeet-tdt-0.6b-v3
  (CoreML conversion by FluidInference)
- OpenAI Whisper large-v3 turbo — MIT — https://github.com/openai/whisper
  (CoreML conversion by Argmax, argmaxinc/whisperkit-coreml)

Text processing
- NeMo inverse text normalization (text-processing-rs, FluidInference) — Apache-2.0

Text refinement
- Qwen3 1.7B / Qwen3 4B Instruct 2507 (Alibaba Qwen) — Apache-2.0 — https://huggingface.co/Qwen
  (GGUF quantization by Unsloth)

Libraries
- KaTeX (math rendering in class notes) — MIT — https://katex.org
- FluidAudio — Apache-2.0 — https://github.com/FluidInference/FluidAudio
- WhisperKit — MIT — https://github.com/argmaxinc/WhisperKit
- llama.cpp — MIT — https://github.com/ggml-org/llama.cpp

Hinglish add-on (downloaded separately from Settings)
  Oriserve/Whisper-Hindi2Hinglish-Swift — Apache-2.0 — https://huggingface.co/Oriserve/Whisper-Hindi2Hinglish-Swift
  Based on OpenAI Whisper (MIT). Converted to Core ML with argmaxinc/whisperkittools (MIT).