Shadowing Practice: AI agents explained: Chatbots vs. workflows vs. agents - Learn English Speaking with Video

Choose a practice mode

Each mode opens a full-screen practice room.

28 sentences
If sentences are too short or too long, click Edit to adjust them.
1
Let's talk about three layers of memory that your AI agent has.
2
Those layers are state, sessions, and long -term memory.
3
So let's use the travel planning agent as an example.
4
First, state.
5
You can think of state as the agent's working memory.
6
It keeps information the agent needs right now, like remaining budget for your trip or current travel plan.
7
It is like a scratch note that you can put important information on it
8
and pass it among different components in your system.
9
And as the plan changes, the state gets updated too
10
and second sessions a session keeps context of one conversation thread
11
it includes everything the conversation transcript like messages two costs two results
12
and state a session service manages saving
13
and loading the information you could use an in -memory backend for example run for quick
14
and easy context or you can add a persistent database so the session can survive server restarts and continue later.
15
This is really important in the long -running process.
16
Third, long -term memory.
17
Long -term memory carries useful information across conversations.
18
In ADK, we have session service to manage all the sessions and memory service to manage the memory across the sessions.
19
To use them in this trip example, you could tell the agent that you prefer quiet neighborhoods.
20
A session service can store all the conversations you ever talk to the agent.
21
A memory service source and retrieves that information
22
and then brings relevant memories into the context so you can plan it next trip.
23
So memory service can help us manage cross -session context.
24
To summarize, each layer has a different job.
25
State is what the agent is working with right now.
26
Sessions preserve the current conversation.
27
A long -term memory carries useful information into the future.
28
And now your AI is context -aware and ready for production.

Download App

Everything you need to speak fluently

AI PronunciationScore every sentence
IPA PracticeMaster every sound
VocabularyBuild your word bank
Vocab GameLearn while playing

Same Channel

FOR YOU

Recommend Video

Vocabulary and speaking notes for this lesson

This speaking lesson is built on the video “AI agents explained: Chatbots vs. workflows vs. agents”. The speaker keeps coming back to these words: session, memory, agent, context, conversation. This video has 28 sentences and 310 words to shadow. The speech runs for 1:54. The speaker talks at a natural 163 words per minute, close to everyday conversation. 88% of the words are among the 3,000 most common in English, so the vocabulary is easy to follow.

Key vocabulary in this video

12 words from the video worth learning, with pronunciation and meaning:

WordPronunciationMeaning
layer noun/ˈleɪ̯ɚ/A single thickness of some material covering a surface.
server noun/ˈsɝvɚ/A program that provides services to other programs or devices, either in the same computer or over a computer network.
thread noun/ˈθɹɛd/A cord formed by spinning or twisting together textile fibers or filaments into one or more continuous strands, typically used in needlework.
component noun/kəmˈpoʊ.nənt/A smaller, self-contained part of a larger entity. Often refers to a manufactured object that is part of a larger device.
database noun/ˈdeɪtəˌbeɪs/A collection of (usually) organized information in a regular structure, usually but not necessarily in a machine-readable format accessible by a computer.
preserve verb/pɹəˈzɝv/To protect; to keep from harm or injury.
scratch verb/skɹæt͡ʃ/To rub a surface with a sharp object, especially by a living creature to remove itching with nails, claws, etc.
persistent adjective/pəˈsɪstənt/Obstinately refusing to give up or let go.
transcript noun/ˈtɹæn.skɹɪpt/Something which has been transcribed; a writing or composition consisting of the same words as the original; a written copy.
retrieve verb/ɹɪˈtɹiːv/To regain or get back something.
restart verb/ˈɹiːˌstɑɹt/To start again.
summarize verb/ˈsʌməˌɹaɪz/To prepare a summary of (something).

How to practise with this video

  1. Listen to the whole video once without speaking and note the words you do not know.
  2. Start at 0.75× speed, shadow it sentence by sentence, then go back to normal speed once it feels easy.
  3. Record yourself and compare with the original, paying attention to words like layer, server, thread.

What is the Shadowing Technique?

Shadowing is a science-backed language learning technique originally developed for professional interpreter training and popularized by polyglot Dr. Alexander Arguelles. The method is simple but powerful: you listen to native English audio and immediately repeat it out loud — like a shadow following the speaker with just a 1–2 second delay. Unlike passive listening or grammar drills, shadowing forces your brain and mouth muscles to simultaneously process and reproduce real speech patterns. Research shows it significantly improves pronunciation accuracy, intonation, rhythm, connected speech, listening comprehension, and speaking fluency — making it one of the most effective methods for IELTS Speaking preparation and real-world English communication.

Shadowing technique: read the full step-by-step guide →