Restez organisé à l'aide des collections
Enregistrez et classez les contenus selon vos préférences.
Le SDK de l'Assistant Google vous permet d'intégrer les commandes vocales, la compréhension du langage naturel et les fonctionnalités intelligentes de Google à vos idées. Votre projet capture un énoncé (une requête audio vocale comme Qu'y a-t-il sur mon agenda ?), l'envoie à l'Assistant Google et reçoit une réponse audio en plus du texte brut de l'énoncé.
Service Assistant Google
Le service de l'Assistant Google expose une API de bas niveau qui vous permet de manipuler directement les octets audio d'une requête et d'une réponse de l'Assistant. Les liaisons de cette API peuvent être générées pour les langages tels que Node.js, Go, C++ et Java pour toutes les plates-formes compatibles avec gRPC.
Le code de référence est fourni en Python pour la capture audio, la lecture audio et la gestion de l'état de la conversation.
Compatibilité et fonctionnalités
Le tableau suivant récapitule les exigences de compatibilité des plates-formes du service de l'Assistant Google:
Sauf indication contraire, le contenu de cette page est régi par une licence Creative Commons Attribution 4.0, et les échantillons de code sont régis par une licence Apache 2.0. Pour en savoir plus, consultez les Règles du site Google Developers. Java est une marque déposée d'Oracle et/ou de ses sociétés affiliées.
Dernière mise à jour le 2025/07/26 (UTC).
[null,null,["Dernière mise à jour le 2025/07/26 (UTC)."],[[["\u003cp\u003eThe Google Assistant SDK enables the integration of voice control, natural language processing, and Google's intelligence into projects, primarily for experimental and non-commercial purposes.\u003c/p\u003e\n"],["\u003cp\u003eIt offers a low-level API for manipulating audio, accessible through various programming languages, along with Python reference code for audio handling and conversation management.\u003c/p\u003e\n"],["\u003cp\u003eWhile supporting audio input/output and visual HTML5 responses, it lacks hands-free activation, timers/alarms, podcast/news playback, and broadcast messaging capabilities.\u003c/p\u003e\n"],["\u003cp\u003eDevelopers can build projects by leveraging the gRPC APIs, customizing functionality with Actions on Google, and adhering to best practices for privacy, security, and audio configuration.\u003c/p\u003e\n"]]],["The Google Assistant SDK facilitates voice control and natural language understanding in projects, using spoken audio requests and responses. It provides a low-level API via gRPC, supporting multiple languages and platforms. Reference code in Python handles audio capture, playback, and conversation state. Key features include visual output of responses, but exclude hands-free activation, timers, alarms, and media playback. Project development involves building with gRPC APIs, customizing, and then refining using privacy and audio best practices. The SDK can only be used for experimental and non-commercial projects.\n"],null,["# Overview\n\n| **Warning:** You can't launch commercial devices that integrate with the Google Assistant SDK. It's available for experimental and non-commercial uses only.\n\n\u003cbr /\u003e\n\nThe Google Assistant SDK lets you add voice control, natural\nlanguage understanding and Google's smarts to your ideas. Your project captures\nan utterance (a spoken audio request, such as *What's on my calendar?*), sends\nit to the Google Assistant, and receives a spoken audio response in addition to\nthe raw text of the utterance.\n\n### Google Assistant Service\n\nThe Google Assistant Service exposes a low level API that lets you directly manipulate the\naudio bytes of an Assistant request and response. Bindings for this API can be\ngenerated for languages like Node.js, Go, C++, Java for all platforms that\nsupport [gRPC](http://www.grpc.io/about/#osp).\n\nReference code is provided in Python for audio capture, audio playback, and\nconversation state management.\n\n### Compatibility and feature support\n\nThe following table summarizes the platform compatibility requirements of\nthe Google Assistant Service:\n\n\n| Google Assistant service | | |\n|----------------------------------------------------------------------------------|-----------------------------------------------------|---|\n| Supported architectures | All [gRPC platforms](http://www.grpc.io/about/#osp) |\n| Supported languages | All [gRPC languages](http://www.grpc.io/about/#osp) |\n| Hands-free activation (*Ok Google*) | No |\n| Audio capture and playback | Reference code is provided |\n| Conversation state management | Reference code is provided |\n| Timers and alarms | No |\n| Playback of podcasts and news | No |\n| [Broadcast voice messages](https://support.google.com/googlehome/answer/7531913) | No |\n| Visual output (HTML5) of Assistant responses | Yes |\n\n\u003cbr /\u003e\n\nSteps to start building your projects\n-------------------------------------\n\nThe SDK lets you build projects quickly by embedding the Google Assistant\nand then letting you add unique functionality with [Actions on Google](/actions):\n\n1. Build your project [using the gRPC APIs](/assistant/sdk/guides/service/python).\n2. Get ideas for [next steps to customize it](/assistant/sdk/guides/service/python/embed/next-steps).\n3. Polish your project by reading best practices for [privacy and security](/assistant/sdk/guides/service/python/best-practices/privacy-and-security) and [audio configuration](/assistant/sdk/guides/service/python/best-practices/audio)."]]
