Thousands of everyday sounds, organized using machine learning.
googlecreativelab/aiexperiments-drum-machine 排名第 14,082，有 763 个星标。
同类别比较
googlecreativelab/aiexperiments-drum-machine 排名第 14,082，此处有 8 个同类条目；carlrannaberg/claudekit 排名第 14,078。
同类别比较| 名称 | 排名 | 指标 |
|---|
| alexandremendoncaalvaro/CorridorKey-Runtime | 14,072 | - 语言:
- C++
- 星标数:
- 764
- 分叉数:
- 20
- 简介:
- Native AI keying runtime and OFX plugin for DaVinci Resolve, built in collaboration with Corridor Digital.
|
|---|
| chenpipi0807/ComfyUI-Index-TTS | 14,074 | - 语言:
- Python
- 星标数:
- 764
- 分叉数:
- 71
- 简介:
- 使用IndexTTS模型在ComfyUI中实现高质量文本到语音转换的自定义节点。支持中文和英文文本，可以基于参考音频复刻声音特征。
|
|---|
| neondatabase/appdotbuild-agent | 14,077 | - 语言:
- Python
- 星标数:
- 764
- 分叉数:
- 113
- 简介:
- The agent that generates working apps (and maybe some other things)
|
|---|
| carlrannaberg/claudekit | 14,078 | - 语言:
- TypeScript
- 星标数:
- 763
- 分叉数:
- 117
- 简介:
- A toolkit of custom commands, hooks, and utilities for Claude Code
|
|---|
| cboard-org/cboard | 14,099 | - 语言:
- JavaScript
- 星标数:
- 762
- 分叉数:
- 287
- 简介:
- Augmentative and Alternative Communication (AAC) system with text-to-speech for the browser
|
|---|
| texturedesign/texturize | 14,102 | - 语言:
- Python
- 星标数:
- 762
- 分叉数:
- 35
- 简介:
- 🤖🖌️ Generate photo-realistic textures based on source images or (soon) PBR materials. Remix, remake, mashup! Useful if you want to create variations on a theme or elaborate on an existing texture.
|
|---|
| NVIDIA-ISAAC-ROS/isaac_ros_nvblox | 14,103 | - 语言:
- C++
- 星标数:
- 761
- 分叉数:
- 149
- 简介:
- NVIDIA-accelerated 3D scene reconstruction and Nav2 local costmap provider using nvblox
|
|---|
| cbyn/bitpredict | 14,107 | - 语言:
- Python
- 星标数:
- 761
- 分叉数:
- 214
- 简介:
- Machine learning for high frequency bitcoin price prediction
|
|---|
关键参数与规格
关键参数与规格| 参数 | 值 |
|---|
| 语言 | JavaScript |
|---|
| 类别 | Applications |
|---|
| 子类别 | Content creation |
|---|
| 关键词 | Music |
|---|
| 星标数 | 763 |
|---|
| 分叉数 | 117 |
|---|
| 贡献者 | 2 |
|---|
| 创建日期 | 2016-11-10 |
|---|
| 更新日期 | 2026-10-02 |
|---|
| 1 天星标数 | 0 |
|---|
| 7 天星标数 | 0 |
|---|
| 排名 | 14,082 |
|---|
| 上次排名 | 13,952 |
|---|
| 排名变化 | -130 |
|---|
| 7 天星标排名 | 14,281 |
|---|
| 上次 7 天星标排名 | 14,297 |
|---|
| 7 天星标排名变化 | 16 |
|---|
| 总星标排名 | 13,877 |
|---|
| 上次总星标排名 | 13,854 |
|---|
| 总星标排名变化 | -23 |
|---|
| 国家 | USA |
|---|
README 摘录
The Infinite Drum Machine Thousands of everyday sounds, organized using machine learning. About Sounds are complex and vary widely. This experiment uses machine learning to organize thousands of everyday sounds. The computer wasn’t given any descriptions or tags – only the audio. Using a technique called t-SNE, the computer placed similar sounds closer together. You can use the map to explore neighborhoods of similar sounds and even make beats using the drum sequencer. Credits Built by Kyle McDonald, Manny
Tan, Yotam Mann, and friends at Google Creative Lab. Thanks to The Philharmonia Orchestra, London for contributing some sounds to this project. Experiments] ( Overview This application is the visualizer and drum machine from this site. In this repo you will find all of the front-end code which visualizes, plays back and makes beats with the audio samples, though it does not contain any audio files or the t-SNE generated from those audio files. You can do that with
your own samples by following some of the resources below. Installation To build the client-side javascript, first install node and webpack. Then you can install of the dependencies and build the files by typing the following in the terminal: Creating t-SNE Map If you have a large audio dataset, you can convert your sounds to t-SNE map using some of the python scripts found in Kyle McDonald's AudioNotebooks repo. Our t-SNE was made by running the
following scripts: First we collect all of our audio files into a single numpy array using and created one large wave file from all of the audio . Then we generated audio fingerprints and turn the fingerprints into a t-SNE map by running and then . The t-SNE data is put into two TSV files. One has all of the 2D coordinates, and the second has is the t-SNE but projected into 3D space instead of 2D. In our
experiment use the 2D coordinates for the position on the map and the 3D coordinates for the color of the data point. In the folder is some more scripts that we used for processing the audio and t-SNE data for our application. To make these sounds work better as a drum machine, we ran an analysis on all of the sounds comparing them to a traditional kick, snare, open and closed hihat using librosa. This generates a TSV with
values for the "distance" of each of our audio samples from these drum reference sounds. Using , you can create a JSON structure the most similar sounds to these reference sounds. This array is used for picking some random defaults when users first open the experiment and the is used for sorting the rest of the sounds into the 4 drum parts when a user clicks the "random" button with a filter in the search box. To deploy our audio files
and data in our experiment, we ran two additional step: first we separated our very large single wave file into
常见问题
记录了多少星标和分叉？
googlecreativelab/aiexperiments-drum-machine 有 763 个星标和 117 个分叉。
仓库排名是多少？
googlecreativelab/aiexperiments-drum-machine 排名第 14,082。
记录的语言和类别是什么？
googlecreativelab/aiexperiments-drum-machine 有 763 个星标，语言为 JavaScript，类别为 Applications。
哪个仓库排名最接近？
googlecreativelab/aiexperiments-drum-machine 与 carlrannaberg/claudekit 比较，后者排名第 14,078，有 763 个星标。
The Infinite Drum Machine
Thousands of everyday sounds, organized using machine learning.
About
Sounds are complex and vary widely. This experiment uses machine learning to organize thousands of everyday sounds. The computer wasn’t given any descriptions or tags – only the audio. Using a technique called t-SNE, the computer placed similar sounds closer together. You can use the map to explore neighborhoods of similar sounds and even make beats using the drum sequencer.
https://aiexperiments.withgoogle.com/drum-machine
This is not an official Google product.
Credits
Built by Kyle McDonald, Manny Tan, Yotam Mann, and friends at Google Creative Lab. Thanks to The Philharmonia Orchestra, London for contributing some sounds to this project. Check out more at [A.I. Experiments] (https://aiexperiments.withgoogle.com).
Overview
This application is the visualizer and drum machine from this site. In this repo you will find all of the front-end code which visualizes, plays back and makes beats with the audio samples, though it does not contain any audio files or the t-SNE generated from those audio files. You can do that with your own samples by following some of the resources below.
Installation
To build the client-side javascript, first install node and webpack. Then you can install of the dependencies and build the files by typing the following in the terminal:
npm install
webpack -p
Creating t-SNE Map
If you have a large audio dataset, you can convert your sounds to t-SNE map using some of the python scripts found in Kyle McDonald's AudioNotebooks repo. Our t-SNE was made by running the following scripts:
First we collect all of our audio files into a single numpy array using Collect Samples.ipynb and created one large wave file from all of the audio Samples to Audio Spritesheet.ipynb. Then we generated audio fingerprints and turn the fingerprints into a t-SNE map by running Samples to Fingerprints.ipynb and then Fingerprint to t-SNE.ipynb.
The t-SNE data is put into two TSV files. One has all of the 2D coordinates, and the second has is the t-SNE but projected into 3D space instead of 2D. In our experiment use the 2D coordinates for the position on the map and the 3D coordinates for the color of the data point.
In the scripts folder is some more scripts that we used for processing the audio and t-SNE data for our application. To make these sounds work better as a drum machine, we ran an analysis on all of the sounds comparing them to a traditional kick, snare, open and closed hihat using librosa. You can see this script in scripts/analysis.py. This generates a TSV with values for the "distance" of each of our audio samples from these drum reference sounds. Using scripts/sort_sounds.py, you can create a JSON structure the most similar sounds to these reference sounds. This array is used for picking some random defaults when users first open the experiment and the analysis.tsv is used for sorting the rest of the sounds into the 4 drum parts when a user clicks the "random" button with a filter in the search box.
To deploy our audio files and data in our experiment, we ran two additional step: first we separated our very large single wave file into many small mp3s to make loading faster and so that it was playable even if all of the sounds weren't loaded yet. And second we created matching JSON meta files for each of the audio chunks which included the position data, color data, analysis data and meta data (this is specific to our audio set which also included a lot of meta data tags).
License
Copyright 2016 Google Inc. All Rights Reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.