diff --git a/CMakeLists.txt b/CMakeLists.txt
index ed88e69..c772c03 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -65,6 +65,11 @@ _install(DIRECTORY "${CMAKE_CURRENT_SOURCE_DIR}/tevr-asr-data/" DESTINATION "/us
 set(CPACK_GENERATOR "DEB")
 set(CPACK_DEBIAN_PACKAGE_MAINTAINER "Hajo Nils Krabbenhöft")
 set(CPACK_DEBIAN_PACKAGE_SHLIBDEPS OFF)
-set(CPACK_PACKAGE_DESCRIPTION "German speech recognition with wav2vec 2.0 XLS-R 1B enhanced with TEVR tokens" )
+set(CPACK_PACKAGE_VERSION "1.0.0")
+set(CPACK_PACKAGE_VERSION_MAJOR 1)
+set(CPACK_PACKAGE_VERSION_MINOR 0)
+set(CPACK_PACKAGE_VERSION_PATCH 0)
+set(CPACK_PACKAGE_FILE_NAME "tevr_asr_tool-1.0.0-Linux-x86_64")
+set(CPACK_PACKAGE_DESCRIPTION "German speech recognition with TEVR as a command-line tool" )
 set(CPACK_DEBIAN_PACKAGE_DEPENDS "libc6 (>= 2.27), libgcc1 (>= 1:4.7), libstdc++6 (>= 6)" )
 include(CPack)
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..de87c78
--- /dev/null
+++ b/README.md
@@ -0,0 +1,121 @@
+# TEVR ASR Tool
+
+* state-of-the-art performance
+* no GPU needed
+* 100% offline
+* 100% private
+* 100% free
+* MIT license
+* Linux x86_64
+* command-line tool
+* easy to understand
+  * only 284 lines of C++ code
+  * AI model on HuggingFace
+
+In August 2022, we ranked 
+**#1 on "Speech Recognition on Common Voice German (using extra training data)"**.
+Accordingly, the performance of this tool is considered to be
+the best of what's currently possible
+in German speech recognition:
+[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/tevr-improving-speech-recognition-by-token/speech-recognition-on-common-voice-german)](https://paperswithcode.com/sota/speech-recognition-on-common-voice-german?p=tevr-improving-speech-recognition-by-token)
+
+## Install the Debian/Ubuntu package
+Download `tevr_asr_tool-1.0.0-Linux-x86_64.deb` from GitHub:
+```bash
+wget "URL_HERE"
+```
+Install it:
+```bash
+sudo dpkg -i tevr_asr_tool-1.0.0-Linux-x86_64.deb
+```
+
+## Install from Source Code
+Download submodules:
+```bash
+git submodule update --init
+```
+CMake configure and build:
+```bash
+cmake -DCMAKE_BUILD_TYPE=MinSizeRel -DCPACK_CMAKE_GENERATOR=Ninja -S . -B build
+cmake --build build --target tevr_asr_tool -j 16
+```
+Create debian package:
+```bash
+(cd build && cpack -G DEB)
+```
+Install it:
+```bash
+sudo dpkg -i build/tevr_asr_tool-1.0.0-Linux-x86_64.deb
+```
+
+## Usage
+
+```bash
+tevr_asr_tool --target_file=test_audio.wav 2>log.txt
+```
+should display the correct transcription
+` mückenstiche sollte man nicht aufkratzen `.
+And `log.txt` will contain the diagnostics and progress 
+that was logged to stderr during execution.
+
+## GPU Acceleration for Developers
+
+I plan to release a Vulkan & OpenGL-accelerated 
+real-time low-latency transcription 
+software for developers soon.
+It'll run 100% private + 100% offline 
+just like this tool,
+but instead of processing a WAV file on CPU
+it'll stream the real-time GPU transcription 
+of your microphone input
+through a WebRTC-capable REST API
+so that you can easily integrate it 
+with your own voice-controlled projects.
+For example, that'll enable 
+hackable voice typing 
+together with `pynput.keyboard`.
+
+If you want to get notified when it launches,
+please enter your email at
+https://madmimi.com/signups/f0da3b13840d40ce9e061cafea6280d5/join
+
+## Commercial / GPU Acceleration
+
+If you have a commercial use-case for this or similar 
+technology - ideally something that helps 
+small and medium-sized businesses in northern Germany
+become more competitive - 
+then please contact me at moin@DeutscheKI.de
+
+
+## Research Citation
+
+If you use this for research, please cite:
+```bibtex
+@misc{https://doi.org/10.48550/arxiv.2206.12693,
+  doi = {10.48550/ARXIV.2206.12693},
+  url = {https://arxiv.org/abs/2206.12693},
+  author = {Krabbenhöft, Hajo Nils and Barth, Erhardt},  
+  keywords = {Computation and Language (cs.CL), Sound (cs.SD), Audio and Speech Processing (eess.AS), FOS: Computer and information sciences, FOS: Computer and information sciences, FOS: Electrical engineering, electronic engineering, information engineering, FOS: Electrical engineering, electronic engineering, information engineering, F.2.1; I.2.6; I.2.7},  
+  title = {TEVR: Improving Speech Recognition by Token Entropy Variance Reduction},  
+  publisher = {arXiv},  
+  year = {2022}, 
+  copyright = {Creative Commons Attribution 4.0 International}
+}
+```
+
+## Replace the AI Model
+
+The German AI model and my training scripts can be found on HuggingFace:
+https://huggingface.co/fxtentacle/wav2vec2-xls-r-1b-tevr
+
+The model has undergone XLS-R cross-language pre-training.
+You can directly fine-tune it with a different 
+language dataset - for example CommonVoice English - 
+and then re-export the files in the
+`tevr-asr-data` folder.
+
+Alternatively, you can donate roughly 2 weeks of 
+A100 GPU credits to me 
+and I'll train a suitable recognition model
+and upload it to HuggingFace.