home ● blog ● projekte ● termine ● internal ● impressum ● datenschutz

Trace:

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
projects:verbal_morality_statute [2024/01/05 21:20] – [Development] kratenkoprojects:verbal_morality_statute [2024/01/05 21:34] (current) – Overhaul kratenko
Line 1: Line 1:
 ====== Verbal Morality Statute Enforcer 2000 ====== ====== Verbal Morality Statute Enforcer 2000 ======
  
-**Documentation is WIP @ 37c3 - please consult nemo / rey / kratenko for questions**+ 
 +<WRAP column 47%> 
 +{{:projects:vmse2000-frontal.jpg|}} 
 +</WRAP> 
 +<WRAP column 47%> 
 +{{:projects:vmse2000-fine.jpg|}} 
 +</WRAP>
  
 The VMSE 2000 is the newest iteration in a long standing series of verbal hate crime prevention devices. It is able to detect language violations in all languages for the region of purchase and works in a range of up to 6 meters, while being able to work in conjunction with other instances of the VMSE2000 to cover your available space and keep you safe from dreaded language violations. The VMSE 2000 is the newest iteration in a long standing series of verbal hate crime prevention devices. It is able to detect language violations in all languages for the region of purchase and works in a range of up to 6 meters, while being able to work in conjunction with other instances of the VMSE2000 to cover your available space and keep you safe from dreaded language violations.
Line 15: Line 21:
   * The public display of the maniacs moral misdemeanour will apply social pressure on the maniac, leading to an adjustment of the subjects moral values.   * The public display of the maniacs moral misdemeanour will apply social pressure on the maniac, leading to an adjustment of the subjects moral values.
   * The continuous enforcement of the verbal morality statute by the VMSE 2000 will result in a better society for everyone's benefit.   * The continuous enforcement of the verbal morality statute by the VMSE 2000 will result in a better society for everyone's benefit.
 +  * The VMSE 2000 emits a modern aura of morality with aesthetics inspired by the best designers of San Angeles.
  
  
Line 23: Line 30:
 ====== Hardware ====== ====== Hardware ======
  
-- Raspberry Pi 5 (4 also successfully tested) +  * Raspberry Pi 5 (4 also successfully tested) 
-- Thermal Printer (compatible with `python-escpos`) +  * Thermal Printer (compatible with `python-escpos`) 
-- USB Audio Adapter +  * USB Audio Adapter 
-- PlayStation Eye USB camera for (taped-over CCD)+  * PlayStation Eye USB camera for (taped-over CCD)
  
 ====== Software used ====== ====== Software used ======
  
-- OpenAI whisper model (base) prompted for your language of choice! +  * OpenAI whisper model (base) prompted for your language of choice! 
-- [[https://github.com/aarnphm/whispercpp|whispercpp python bindings]] +  * [[https://github.com/aarnphm/whispercpp|whispercpp python bindings]] 
-- [https://github.com/openvinotoolkit/openvino|OpenVINO]] for speeding up encoding+  * [[https://github.com/openvinotoolkit/openvino|OpenVINO]] for speeding up encoding
  
 ====== Source Code ====== ====== Source Code ======
  
-You can find the firmware / device glue at [[https://github.com/deepestcyber/vmse2000-firmware/]]. +  * You can find the firmware / device glue at [[https://github.com/deepestcyber/vmse2000-firmware/]]. 
-The voice detection part can be found at [[https://github.com/deepestcyber/vmse2000-detector/]].+  * The voice detection part can be found at [[https://github.com/deepestcyber/vmse2000-detector/]].
  
-===== Previous Iterations =====+====== Previous Iterations ======
  
 There were a lot of iterations to get to this result (and a lot of non-development as well - this project started in 2017 after all). There were a lot of iterations to get to this result (and a lot of non-development as well - this project started in 2017 after all).
 We tested DeepSpeech, DeepSpeech V2, RNN on DeepSpeech 2 feature extractors and binary classification RNNs trained from scratch. In the end the simplest and most robust model was OpenAI whisper. Our suspicion is that the amount of data, it's variance and the resulting robustness to noise (microphone as well as background) is what makes the difference. We tested DeepSpeech, DeepSpeech V2, RNN on DeepSpeech 2 feature extractors and binary classification RNNs trained from scratch. In the end the simplest and most robust model was OpenAI whisper. Our suspicion is that the amount of data, it's variance and the resulting robustness to noise (microphone as well as background) is what makes the difference.
- 
- 
- 
-===== Material needed ===== 
- 
-==== Data ==== 
-  * [[https://tatoeba.org/|Tatoeba]] hat Daten in verschiedenen Sprachen mit [[https://tatoeba.org/eng/tags/show_sentences_with_tag/454|tags]] und auch Audio 
-  * https://gist.github.com/jamiew/1112488 -- All the dirty words from Google's "what do you love" project: http://www.wdyl.com/ 
-  * http://www.cs.cmu.edu/~biglou/resources/bad-words.txt -- another bad word list 
-  * https://www.hatebase.org/ -- world's largest online repository of structured, multilingual, usage-based hate speech 
-  * https://github.com/t-davidson/hate-speech-and-offensive-language -- ein paper dazu (textbasiert) 
-  * //Sentiment analyses of single words or short phrases// unter https://www.crowdflower.com/data-for-everyone/ 
-==== SW ==== 
- 
-  * daten! big! für deepest cyber! 
-  * ASR 
-     * https://github.com/PaddlePaddle/DeepSpeech 
-     * https://github.com/mozilla/DeepSpeech 
- 
-==== HW ==== 
-  * two area micros -> rey fragt mal rum 
-  * two pairs of speakers -> rey fragt mal rum 
-  * amplifier for speakers -> rey fragt mal rum 
-  * two sound cards for Raspi (USB) 
-  * suitable cases 
- 
-Bonus: 
- 
-  * portable power (printer needs a lot of power) 
  
  
  
  
Last modified: 2024/01/05 21:34