{"id":13630235,"url":"https://github.com/jaxcore/bumblebee","last_synced_at":"2025-04-17T13:31:50.478Z","repository":{"id":37181688,"uuid":"273362244","full_name":"jaxcore/bumblebee","owner":"jaxcore","description":"Jaxcore Bumblebee - a JavaScript voice application framework","archived":false,"fork":false,"pushed_at":"2023-03-06T17:30:45.000Z","size":13223,"stargazers_count":429,"open_issues_count":31,"forks_count":21,"subscribers_count":24,"default_branch":"master","last_synced_at":"2025-04-06T16:09:26.946Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"JavaScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/jaxcore.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2020-06-18T23:44:55.000Z","updated_at":"2024-12-17T08:12:14.000Z","dependencies_parsed_at":"2024-08-01T22:52:24.430Z","dependency_job_id":null,"html_url":"https://github.com/jaxcore/bumblebee","commit_stats":{"total_commits":129,"total_committers":1,"mean_commits":129.0,"dds":0.0,"last_synced_commit":"1a39be6750a25fb938284b2ab8f6ef2f1b30f363"},"previous_names":[],"tags_count":3,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jaxcore%2Fbumblebee","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jaxcore%2Fbumblebee/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jaxcore%2Fbumblebee/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jaxcore%2Fbumblebee/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/jaxcore","download_url":"https://codeload.github.com/jaxcore/bumblebee/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":249337342,"owners_count":21253529,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-08-01T22:01:35.114Z","updated_at":"2025-04-17T13:31:48.698Z","avatar_url":"https://github.com/jaxcore.png","language":"JavaScript","funding_links":[],"categories":["JavaScript"],"sub_categories":[],"readme":"![screenshot](assets/logo.png)\n\n### JavaScript Voice Application Framework\n\nWrite your own voice apps and assistants with an easy-to-learn JavaScript API!\n\n\n## About\n\nBumblebee is a set of libraries, tools, and methodologies that enables JavaScript developers to write their own conversational voice assistants in either NodeJS or on the web.\n\nThe open source technologies that Bumblebee uses are:\n\n- [NodeJS](https://nodejs.org/en/) - V8 JavaScript engine\n- [ElectronJS](https://www.electronjs.org/) - JavaScript desktop application framework\n- [Mozilla DeepSpeech](https://github.com/mozilla/DeepSpeech) - Tensorflow-based speech-to-text processing\n- [Picovoice Porcupine](https://github.com/Picovoice/porcupine) - JavaScript hotword detection\n- [meSpeak](https://www.masswerk.at/mespeak/) - JavaScript text-to-speech library\n\nBumblebee builds upon these technologies by making them easier to use and tying them\ntogether into an integrated voice application API.  Because these systems\nrun locally without any cloud services, stand-alone privacy-focused always-on voice applications can\nfinally be realized.\n\nThe Bumblebee project includes an ElectronJS voice app server console,\nwhich automatically installs and sets up DeepSpeech, and runs the bumblebee websocket service that\nvoice applications can connect to. The applications run independently of the bumblebee server,\nand use a websocket API to connect and communicate by receiving speech-to-text results\nand hotword commands, or issuing text-to-speech instructions.\n\nThere are limitless ways to expand Bumblebee's capabilities by writing new applications that control\ndevices or services on a home network, retrieve data from the internet, and anything else you can think of.\nAnd as you will see, they are both ***EASIER*** and ***MORE FUN*** to write than you think.\n\nBumblebee voice apps are small, simple, single file scripts. They can be shared easily between\nsystems and build upon eachother to create larger and smarter voice applications.\nBy associating a voice app with a hotword, your app becomes a voice assistant\nthat can be called upon at any time.\n\n## Installation\n\n### Requirements\n\nThe computing resources required to run Bumblebee are much larger than a typical application.\n\n- Disk Space: 1.9 GB of space is required for installation\n\t- Bumblebee: ~450 MB\n\t- DeepSpeech: ~1.4 GB\n- CPU's with AVX support are [required](https://github.com/tensorflow/tensorflow/issues/19584), GPU may be utilized if available\n- RAM usage will fluctuate ~350 MB (or higher)\n\n### Desktop Application\n\nTo install and run the Bumblebee desktop app from the source files, see:\n\n- [INSTALL.md](electron-app/INSTALL.md)\n\nAfter installation, the first time Bumblebee is run it will prompt to download the DeepSpeech [pre-trained English model](https://github.com/mozilla/DeepSpeech/releases) files (1.4 GB disk space required).\n\nWhen installed successfully and with the microphone and speakers turned on, the console will react in real-time to the audio it hears.  Test the that speech-to-text is operational before proceeding.\n\n![screenshot](assets/screenshot.png)\n\n## Hello World\n\nTo get started, create the most simple Bumblebee app possible, a \"Hello World\" voice application.\n\nCreate a new directory and NPM project:\n\n```\nmkdir helloworld\ncd helloworld\nnpm init\nnpm install jaxcore-bumblebee\n```\n\nCreate a new file named `helloworld.js`:\n\n```\nconst Bumblebee = require('jaxcore-bumblebee');\n\nclass HelloWorldApp extends Bumblebee.Application {\n\tconstructor() {\n\t\tsuper(...arguments);\n\t}\n\n\tasync loop() {\n\t\tthis.console('Say \"Hello World\"');\n\n\t\tlet recognition = await this.recognize();\n\t\tthis.console(recognition);\n\n\t\tif (recognition.text === 'hello world') {\n\t\t\tawait this.playSound('okay');\n\t\t\tawait this.say('Hello World');\n\t\t}\n\t\telse {\n\t\t\tawait this.playSound('error');\n\t\t}\n\t}\n}\n\nBumblebee.connectApplication(HelloWorldApp, {\n\tname: \"Hello World\",\n\tautoStart: true\n});\n```\n\nRun the voice app:\n\n```\nnode helloworld.js\n```\n\nWith the speakers and microphone turned on, you will be able to talk to this application and listen to it's responses.  It doesn't do very much yet, if it hears you say \"hello world' it will respond by making a beep sound and also saying \"hello world\".\n\nThis program will run continuously until the NodeJS script is closed using `Command+C` or `Control+C` and it can be started and stopped at any time.\n\n## Documentation (Coming Soon)\n\nThis is a brand new project with much more to come.\n\nThere are some additional [examples](https://github.com/jaxcore/bumblebee/tree/master/examples) to try and many more are in the works.\n\nUse github's \"watch\" feature to stay tuned for updates!","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjaxcore%2Fbumblebee","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjaxcore%2Fbumblebee","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjaxcore%2Fbumblebee/lists"}