Differences
This shows you the differences between two versions of the page.
| Both sides previous revisionPrevious revisionNext revision | Previous revision | ||
| projects:verbal_morality_statute [2018/10/07 17:44] – kratenko | projects: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 ====== | ||
| - | ===== Ways to go ===== | ||
| - | * snips als grundlage ausprobieren? | ||
| - | ===== Open tasks ===== | + | <WRAP column 47%> |
| - | * deep speech testen (all) | + | {{: |
| - | * deepspeech (mozilla) auf continuous umbauen | + | </WRAP> |
| - | * deepspeech (paddlepaddle) ausprobieren | + | <WRAP column 47%> |
| - | * validation set baun | + | {{: |
| - | * corpora | + | </ |
| - | * librispeech | + | |
| - | * wort/phrase liste anlegen | + | |
| - | * Ger | + | |
| - | * Eng | + | |
| - | * Ausgabe lösen (wav nehmen? odertts?) | + | |
| - | * Beleuchtung baun (blink-a-lot) | + | |
| - | * Gehäuse (Wandanbringing) | + | |
| - | * speaker + amplifier + power supply | + | |
| - | * Stromversorgung | + | |
| + | 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. | ||
| - | ------ | + | Our analysts predict that the device integrate into everyone' |
| - | ===== General Function ===== | ||
| - | |||
| - | The VMSE2000 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. | ||
| ===== Detailed Specification ===== | ===== Detailed Specification ===== | ||
| - | * The VMSE2000 | + | * The VMSE 2000 listens for speech input and detects |
| - | * Consequences of a detected language violation is a verbal notification as well as a printed receipt for a credit fine of 1$ in BTC including | + | * Consequences of a detected language violation is a verbal notification as well as a printed receipt for a credit fine of a sum adequate |
| - | * The language | + | * The morality |
| + | * 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' | ||
| + | * The VMSE 2000 emits a modern aura of morality with aesthetics inspired by the best designers of San Angeles. | ||
| - | ===== What needs to be done ===== | ||
| - | Minimum Viable Product: | + | ===== Development ===== |
| - | * evaluate Kaldi | + | The VMSE 2000 is one of the most important projects by Deep Cyber, if not one of the most important efforts of our lifetime. While the physical parameters and technical details |
| - | * does it still have pre-trained models? | + | |
| - | * does it run on a Raspi? | + | |
| - | * find alternatives to Kaldi | + | |
| - | * data set of language | + | |
| - | * dict.cc | + | |
| - | * leo | + | |
| - | * movie dataset? | + | |
| - | * there should | + | |
| - | * ensure passable noise robustness | + | |
| - | ===== Design | + | ====== Hardware ====== |
| + | * Raspberry Pi 5 (4 also successfully tested) | ||
| + | * Thermal Printer (compatible with `python-escpos`) | ||
| + | * USB Audio Adapter | ||
| + | * PlayStation Eye USB camera for (taped-over CCD) | ||
| - | ===== Material needed | + | ====== Software used ====== |
| - | ==== Data ==== | + | * OpenAI whisper model (base) prompted for your language of choice! |
| - | * [[https://tatoeba.org/ | + | * [[https://github.com/aarnphm/whispercpp|whispercpp python bindings]] |
| - | * https://gist.github.com/jamiew/1112488 -- All the dirty words from Google' | + | * [[https:// |
| - | * http:// | + | |
| - | * https:// | + | |
| - | * https:// | + | |
| - | * //Sentiment analyses of single words or short phrases// unter https:// | + | |
| - | ==== SW ==== | + | |
| - | * daten! big! für deepest cyber! | + | ====== Source Code ====== |
| - | * ASR | + | |
| - | * https:// | + | |
| - | * https:// | + | |
| - | ==== HW ==== | + | |
| - | | + | * The voice detection part can be found at [[https:// |
| - | * two pairs of speakers | + | |
| - | * amplifier for speakers -> rey fragt mal rum | + | |
| - | * two sound cards for Raspi (USB) | + | |
| - | * suitable cases | + | |
| - | Bonus: | + | ====== Previous Iterations ====== |
| - | * portable power (printer needs a lot of power) | + | There were a lot of iterations to get to this result |
| + | 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. | ||
Last modified: 2018/10/07 17:44