Languagez Licenses

Last updated: July 30, 2026

Languagez is developed and published by Innobytez.

This page provides third-party software notices and model attribution for the current mobile release line.

This is an attribution and review record, not a grant of rights. Two sherpa model families used by the current Android build do not have complete, publicly verifiable license terms: the Kroko English, Spanish, French, and German bundles, and the gated Chinese source checkpoint. They should not be treated as cleared for commercial distribution until the applicable written terms are obtained and retained.

Overview

Languagez includes:

Flutter framework/runtime and Dart packages.
Native speech recognition components (whisper.cpp and sherpa-onnx).
Platform media/PDF libraries on Android.
Downloadable speech model files used at runtime.

Third-party software components

Component	Used in	License	Source
Flutter SDK and Dart SDK	iOS + Android	BSD-3-Clause and package-specific	https://github.com/flutter/flutter
Flutter/Dart packages from `pubspec.yaml`	iOS + Android	Package-specific (see in-app licenses)	https://pub.dev
whisper.cpp / ggml	iOS + Android transcription engine	MIT	https://github.com/ggerganov/whisper.cpp
sherpa-onnx engine	Android live ASR engine	Apache-2.0	https://github.com/k2-fsa/sherpa-onnx
Google ML Kit translation and language identification	iOS + Android on-device translation	Google ML Kit Terms; translation usage and attribution requirements also apply	ML Kit Terms and translation usage guidelines
AndroidX Media3 (`media3-transformer`, `media3-common`, `media3-effect`)	Android media processing	Apache-2.0	https://github.com/androidx/media
pdfbox-android (`com.tom-roush:pdfbox-android`)	Android PDF processing	Apache-2.0	https://github.com/TomRoush/PdfBox-Android

Downloadable model assets

These models are optional and downloaded on demand; they are not all active at once. Model licensing can be different from engine licensing. The following model sources have reachable runtime paths in the current app:

When each model is used

Whisper ggml-small.bin transcribes imported audio and video on iOS and Android.
The English, Spanish, French, German, and Chinese sherpa streaming bundles transcribe Android live audio when that source language is selected or remembered.
Omnilingual ASR and Silero VAD work together for Android live audio when Auto has no supported language hint or when Italian, Portuguese, Hindi, or Japanese is selected. Silero segments speech; Omnilingual transcribes those segments.
Google ML Kit translation models perform on-device translation on iOS and Android. ML Kit language identification handles Auto source-language detection.

Asset	Source URL	Current upstream license signal	Status
Whisper model (`ggml-small.bin`)	https://huggingface.co/ggerganov/whisper.cpp	MIT	Verified; the app pins SHA-256 `1be3a9b2063867b937e64e2ec7483364a79917e157fa98c5d94b5c1fffea987b`
Sherpa streaming bundles (en/es/fr/de)	English, Spanish, French, and German	The bundle README points to Banafo/Kroko-ASR. Its model card describes community models as CC-BY-SA and separate commercial models, but gives no CC version. The repository metadata says `License: test`, and its formal LICENSE file is empty.	Unresolved — do not treat as cleared for commercial distribution
Sherpa streaming bundle (zh)	https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-streaming-zipformer-zh-int8-2025-06-30.tar.bz2	The converted model page has no license metadata and points to a gated source checkpoint whose conditions are visible only after sign-in and acceptance.	Unresolved — obtain and retain the gated terms before distribution
Sherpa omnilingual bundle	https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-omnilingual-asr-1600-languages-300M-ctc-v2-int8-2026-02-05.tar.bz2	Apache-2.0. The current downloadable archive contains Meta's Apache-2.0 `LICENSE`. The conversion model card identifies the conversion, but its displayed size and checksum for this filename are older than the current GitHub release asset.	Verified against the app pin and current GitHub release asset; SHA-256 `951b32409aade32bd525310bb39e9666773ba3fc611a39e817f620936d76c631`
Silero VAD (`silero_vad.onnx`)	https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/silero_vad.onnx	MIT under the upstream Silero VAD license	Upstream license verified; current release mirror SHA-256 `9e2449e1087496d8d4caba907f23e0bd3f78d91fa552479bb9c23ac09cbb1fd6`
Google ML Kit downloadable translation models	Downloaded and managed by Google ML Kit; see the on-device translation documentation	Governed as related software by the Google ML Kit Terms rather than published under an open-source model license	Terms identified; confirm the app's required Google Translate attribution and branding before release

License texts

whisper.cpp / ggml MIT License

Copyright (c) 2023-2026 The ggml authors

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.

Apache License 2.0

The following components use Apache-2.0:

sherpa-onnx engine
AndroidX Media3
pdfbox-android
the pinned omnilingual ASR model archive identified above

Full text: https://www.apache.org/licenses/LICENSE-2.0

Other Terms And Notices

Silero VAD uses the Silero Team MIT License.
Google ML Kit and its downloadable models are governed by the Google ML Kit Terms. The on-device Translation API also has usage and attribution requirements.

Contact

If you need complete third-party attribution details for a specific app build, contact: [email protected].