Key takeaways
- Gemini Live is a back-and-forth voice conversation with Gemini that you can interrupt, and it can see through your camera or screen.
- On Android, Gemini opens from the power button, a corner swipe or 'Hey Google'; on iPhone, from the Gemini app.
- You can switch the default assistant or disable the app on most phones, though options depend on the device.
- For PDFs, upload the file and ask for summaries, explanations and quizzes; check answers against the pages.
- Video generation with Veo depends on your plan and region and has daily limits.
Gemini Live is the real-time voice mode of Google's Gemini assistant: you talk naturally, interrupt whenever you like, and Gemini answers out loud; you can also share your camera or screen so it can see what you are looking at. On Android it is built into the phone's assistant; on iPhone it is in the Gemini app. This page explains how Live works, how to find and use Gemini on your phone, whether you can remove it, and short tutorials for Gemini in general, PDFs, Veo video generation and notebooks.
What is Gemini Live?
Gemini Live is a conversational voice mode where you speak with Gemini in real time, as in a phone call. You can interrupt it, change topic, and ask follow-ups, and it keeps the conversation's context.
Live also lets you share your camera, so Gemini can talk about what it sees, such as a plant, a circuit board or a menu in another language, and your screen, so it can help with what you are looking at. It is useful for practising interviews or spoken English, brainstorming while walking, learning a topic by asking questions aloud, or getting help with something physical in front of you. Transcripts of Live conversations appear in your Gemini history.
How does Gemini Live work?
Gemini Live streams your voice to Google's servers, where a Gemini model built for low-latency audio understands speech and generates a spoken reply, while listening for interruptions. Camera and screen sharing add a live video feed to the conversation.
Because it is designed for conversation, it uses natural turn-taking: it stops when you start speaking and picks up from where you interrupted. It needs an internet connection and uses mobile data, so expect higher data use on video. Voice options are in settings. Developers can build similar real-time voice and video experiences with the Gemini Live API. As with any assistant, check important facts it states aloud, since spoken answers are harder to verify at a glance.
Where is Gemini on my phone?
On most recent Android phones, Gemini is the default assistant: hold the power button, swipe up from a bottom corner, or say "Hey Google". It is also an app called Gemini in your app drawer. On iPhone, open the Gemini app from the App Store.
Also asked as: where's gemini on my phone
If you do not see it on Android, update the Google app and Play Store apps, or install Gemini from Google Play. Some phones, such as certain Samsung models, also add a Gemini shortcut to the side button. On iPhone you can add Gemini to the Lock Screen or use a Shortcut to open it quickly. Make sure you install the official app published by Google LLC; fake "Gemini" apps exist.
How do I use Gemini on an Android phone?
Open Gemini from the power button, a corner swipe or the app, then type, speak or tap the Live button for a voice conversation. Use the camera and screen options to ask about what you see.
Also asked as: how to gemini on an android phone
Things worth trying:
- "Summarise this page" while a long article is on screen.
- Circle to Search on supported phones to ask about anything on screen.
- "Set a reminder to submit the assignment at 6 pm tomorrow."
- Photograph your notes and ask for a summary or quiz.
- Ask about a YouTube video while it plays, where supported.
Connected apps, such as Google Maps, Calendar, Keep and YouTube, let Gemini act on your behalf. Review which apps it can access in Gemini's settings.
How does Gemini work on my phone?
Gemini on your phone sends your requests to Google's Gemini models in the cloud, uses on-device models for some quick features, and connects to apps such as Calendar, Messages, Maps and YouTube when you allow it.
Most answers come from cloud models, so an internet connection is needed. Some phones also run Gemini Nano, a small on-device model, for features such as summarising recordings or smart replies, which work faster and keep data on the device. When you ask Gemini to do something in another app, it uses that app's permissions. Your conversations are saved to your Google account according to your activity settings, which you can change or turn off in Gemini's privacy settings.
Can I remove Gemini from my phone?
On most Android phones you can switch the default digital assistant away from Gemini in settings, and disable or uninstall the Gemini app. Whether you can fully remove it depends on your phone and Android version.
Look in Settings, then Apps, then Default apps, then Digital assistant app, to change the assistant. To stop Gemini opening from the power button, change the power button setting. If the Gemini app came preinstalled as a system app, you may only be able to disable it or uninstall its updates. Google has been replacing Google Assistant with Gemini on Android, so switching back may not be available on every device. On iPhone, simply delete the app.
Is there a Gemini tutorial for beginners?
Yes. Google's Gemini help centre has guides for each feature, and the fastest way to learn is to try five everyday tasks in your first week.
Also asked as: tutorial gemini · tutorial gemini notebook
A beginner's week:
- Day 1: ask Gemini to explain a topic you are studying and quiz you.
- Day 2: upload a PDF and ask for a one-page summary.
- Day 3: use Gemini Live to practise answering interview questions aloud.
- Day 4: create a Gem, a custom assistant with saved instructions, for a task you repeat.
- Day 5: connect Google Calendar and plan your week.
Write prompts with context: who you are, what you need and the format you want. Check facts in every answer.
How do I use Gemini with a PDF?
Upload the PDF in the Gemini app with the plus button, then ask questions: a summary, explanations of difficult sections, key terms, or practice questions. Gemini can read long documents and point you to relevant parts.
Also asked as: tutorial gemini pdf
Good prompts for study:
- "Summarise this chapter in ten bullet points for revision."
- "Explain section 3.2 as if I am new to the topic, with an example."
- "List all definitions in this PDF as a glossary."
- "Write five exam-style questions from this document, then check my answers."
For scanned PDFs, Gemini can read the images, but accuracy drops with poor scans. For courses with many documents, NotebookLM is often better, because it keeps sources together and cites the exact passages.
How do I use Veo 3 in Gemini?
In the Gemini app, choose the video option, describe the scene you want, and Gemini generates a short clip with Google's Veo model, including sound. Access depends on your plan and country, and there are daily limits.
Also asked as: tutorial gemini veo 3
Write prompts like a director: subject, action, setting, camera movement, lighting, style and sound, for example "a slow dolly shot of a student coding at night in a small hostel room, warm desk lamp, rain on the window, soft keyboard sounds". Generate several versions and keep the best. Clips are short, so plan longer videos as several shots edited together. Generated videos carry a visible or invisible watermark; label them as AI-generated when you share them, and never create videos of real people without consent.
How do I use Gemini notebooks?
Notebooks in Google's AI tools usually means NotebookLM: create a notebook, add sources such as PDFs, Docs, websites or YouTube videos, then ask questions answered only from those sources, with citations. Google has also been connecting notebooks to the Gemini app.
A study workflow: create one notebook per subject, add your syllabus, lecture notes and textbook chapters, and ask for a study guide, a list of likely exam questions, and explanations of the hardest topics. Click citations to check each answer against your material. Audio overviews turn your sources into a podcast-style discussion you can listen to while travelling. Keep sources accurate and current, because answers are only as good as what you upload.
Sources
GenAI engineer and AI product builder with 2+ years shipping production-grade AI systems: RAG pipelines, fine-tuned LLMs, hybrid retrieval and multi-modal apps across vision, speech and OCR, architected end to end from ingestion to deployment. Leads engineering for SCULT INDIA's 14-member team, founded the 500+ member TechVerse Enclave community and has mentored 200+ students. Three hackathon first prizes: Changethon 2025 (IIT Roorkee), Product Genesis at Vividhotsava 2025 (CSJMU) and BYTEBATTLE (MeetKats).
Open to GenAI roles, hackathon judging, mentorship sessions and guest talks at colleges. For campus sessions the inviting institution covers the visit end to end: travel (a First AC train ticket for journeys up to five hours, a flight for anything longer), accommodation, meals, and pickup and drop; the honorarium and workshop fees are negotiable on WhatsApp or on the call, and requests are welcome from institutions that can cover the visit this way. Email pranjulrathour41@gmail.com.
