Footer

    Download on the App StoreGet it on Google Play

    關於

    • 認識 VoiceTube
    • 學習服務介紹
    • 加入我們
    • 常見問題
    • 熱門搜尋主題
    • 企業英文培訓
    • 社群推廣分潤計畫

    服務總覽

    • 口說挑戰
    • 單字單句本
    • Hero 智能學習
    • Tutor 真人家教
    • Vclass 名師課程
    • Campus 教育版
    • 字典查詢
    • 匯入影片並生成字幕
    • 部落格

    精選頻道

    影片分級

    • A1 初級
    • A2 初級
    • B1 中級
    • B2 中高級
    • C1 高級
    • C2 高級

    隱私權˙條款˙
    ©2026 VoiceTube Corporation. All rights reserved

    tensor

    US /ˈtɛnsɚ, -ˌsɔr/

    ・

    UK /ˈtensə(r)/

    B2 中高級
    n.名詞張肌 ; 張量
    This tensor allows you to bend your arm

    影片字幕

    Nvidia 2024 年電腦展主題演講,15 分鐘濃縮精華!(Nvidia's 2024 Computex Keynote: Everything Revealed in 15 Minutes)

    14:40Nvidia 2024 年電腦展主題演講,15 分鐘濃縮精華!(Nvidia's 2024 Computex Keynote: Everything Revealed in 15 Minutes)
    • It has to distribute the workload across multiple GPUs, tensor parallelism, pipeline parallelism, data parallelism, all kinds of parallelism, expert parallelism, all kinds of parallelism, distributing the workload across multiple GPUs, processing it as fast as possible because if you are in a factory, if you run a factory, your throughput directly correlates to your revenues.

      但是,運行人工智能的計算堆棧複雜得令人難以置信。

    • It has to distribute the workload across multiple GPUs, tensor parallelism, pipeline parallelism, all kinds of parallelism, expert parallelism, all kinds of parallelism, distributing the workload across multiple GPUs, processing it as fast as possible because if you are in a factory, if you run a factory, your throughput directly correlates to your revenues.

      但是,運行人工智能的計算堆棧複雜得令人難以置信。

    B1 中級

    Amazon 的新 AI 資料中心不需要 NVIDIA 晶片!規模之大令人難以置信! (No Nvidia Chips Needed! Amazon’s New AI Data Center For Anthropic Is Truly Massive)

    16:56Amazon 的新 AI 資料中心不需要 NVIDIA 晶片!規模之大令人難以置信! (No Nvidia Chips Needed! Amazon’s New AI Data Center For Anthropic Is Truly Massive)
    • Its Tensor processing units, first announced in 2016 are used to power Google's own AI model, Gemini.

      它的 Tensor 處理單元(TPU),最早於 2016 年宣佈,用於驅動 Google 自家的 AI 模型 Gemini。

    • Its tensor processing units, first announced in 2016, are used to power Google's own AI model, Gemini.

      沒有。

    B1 中級

    與輝達執行長黃仁勳的對談:微軟 Build 2025 會議上的薩提亞·納德拉! (Conversation with NVIDIA CEO Jensen Huang: Satya Nadella at Microsoft Build 2025)

    12:36與輝達執行長黃仁勳的對談:微軟 Build 2025 會議上的薩提亞·納德拉! (Conversation with NVIDIA CEO Jensen Huang: Satya Nadella at Microsoft Build 2025)
    • It's liquid cooled, the architecture is FP4 tensor core.

      它是液體冷卻的,架構是 FP4 tensor core。

    • The architecture is FP4 tensor core.

      這在短短兩年內就達到了驚人的加速。

    B1 中級

    Pixel 10a 就是 Pixel 9a 嗎?! (The Pixel 10a is the Pixel 9a?)

    13:49Pixel 10a 就是 Pixel 9a 嗎?! (The Pixel 10a is the Pixel 9a?)
    • Uh, chip, same exact Pixel, the Tensor 4 chip.

      而且它跟 Pixel 9a 一模一樣,幾乎字面上來說就是一樣。

    • The Tensor four chip.

      Tensor 4 晶片。

    B1 中級

    粉紅 Nothing 4a 激夯! (The Pink Nothing 4a is HOT)

    16:01粉紅 Nothing 4a 激夯! (The Pink Nothing 4a is HOT)
    • I use a Tensor chip so clearly I don't give a about performance but I think like what Marques said in like this more range of things when you're probably cross shopping if I'm looking at two phones at the same price and one chip is better then I'm probably going to go that for the longevity route.

      我用 Tensor 晶片,所以很明顯我不在乎效能,但我認為就像 Marques 說的,在比較多樣化的東西裡,當你在交叉比對時,如果我看到兩支價格一樣的手機,其中一支晶片比較好,那我可能會選擇那支,為了長久使用。

    • I use a Tensor chip, so clearly I don't give a shit about performance.

      我覺得這就是那兩個不同族群的完美例子。

    B1 中級

    Windows 筆記型電腦到底怎麼了?🤔 (What's Going on with Windows Laptops?)

    17:12Windows 筆記型電腦到底怎麼了?🤔 (What's Going on with Windows Laptops?)
    • Now Tensor.

      而 Microsoft 真的,他們試圖透過 Surface 進行垂直整合,

    • Now Tensor.

      現在是 Tensor。

    B1 中級

    顯示卡怎麼運作?帶你深入探索 GPU 架構! (How do Graphics Cards Work? Exploring GPU Architecture)

    28:30顯示卡怎麼運作?帶你深入探索 GPU 架構! (How do Graphics Cards Work?  Exploring GPU Architecture)
    • Next, inside each of these streaming multiprocessors are four warps and one ray tracing core, and then inside each warp are 32 CUDA, or shading cores, and one tensor core.

      接下來,每個流式多處理器內有四個 warp 和一個光線追蹤核心,每個 warp 內有 32 個 CUDA(即著色核心)和一個張量核心。

    • Across the entire GPU are 10,752 CUDA cores, 336 tensor cores, and 84 ray tracing cores.

      整個 GPU 擁有 10,752 個 CUDA 內核、336 個張量內核和 84 個光線追蹤內核。

    B2 中高級

    Google Pixel 10A 實機初體驗:前所「未」見的驚喜!🤯 (Google Pixel 10A Impressions: (Never) Seen This Before)

    08:31Google Pixel 10A 實機初體驗:前所「未」見的驚喜!🤯 (Google Pixel 10A Impressions: (Never) Seen This Before)
    • And inside it's even the same tensor G4 chip from last year.

      而且內部一樣是去年的 Tensor G4 晶片。

    • And inside, it's even the same Tensor G4 chip from last year, so it's not even a spec bump.

      所以這甚至不是 spec bump。

    B1 中級

    Google Pixel 9A 搶先體驗:這項超強規格太驚人! (Google Pixel 9A Impressions: One Great Spec!)

    06:50Google Pixel 9A 搶先體驗:這項超強規格太驚人! (Google Pixel 9A Impressions: One Great Spec!)
    • It is powered by the same Tensor G4 chip as the flagship Pixel 9s, with eight gigs of RAM and 128 gigs of starting storage.

      它採用了與旗艦產品 Pixel 9s 相同的 Tensor G4 芯片,配備 8GB 內存和 128GB 啟動存儲空間。

    • It is powered by the same Tensor G4 chip as the flagship Pixel 9s with 8 gigs of RAM and 128 gigs of starting storage,

      它有扁平的鋁製導軌。

    B1 中級

    NVIDIA Vera Rubin 深度解析 — 比 Blackwell 更強大的 10 倍效率接班人! (Deconstructing Nvidia’s Vera Rubin — The Successor To Blackwell That’s 10x More Efficient)

    14:00NVIDIA Vera Rubin 深度解析 — 比 Blackwell 更強大的 10 倍效率接班人! (Deconstructing Nvidia’s Vera Rubin — The Successor To Blackwell That’s 10x More Efficient)
    • And Google's filled its data centers with racks of its own Tensor processing units, too.

      對吧?

    • And Google's filled its data centers with racks of its own Tensor processing units too.

      Google 也用自家 Tensor processing units 的機架填滿了它的 data centers。

    B1 中級