> ## Documentation Index
> Fetch the complete documentation index at: https://dragonwingdocs.qualcomm.com/llms.txt
> Use this file to discover all available pages before exploring further.

# 术语表

> 针对 Dragonwing IoT 平台的 Qualcomm AI/ML 开发者工作流文档中所使用的缩略语、工具和术语的定义。

| **缩略语/术语**   | **说明**                                                       |
| ------------ | ------------------------------------------------------------ |
| AI           | 人工智能 (Artificial Intelligence)                               |
| ANN          | Android 神经网络 (Android Neural Network)                        |
| API          | 应用程序编程接口 (Application Programming Interface)                 |
| ASR          | 自动语音识别 (Automatic Speech Recognition)                        |
| BE           | 后端 (Back End)                                                |
| CPU          | 中央处理单元 (Central Processing Unit)                             |
| CSV          | 逗号分隔值 (Comma-Separated Value)                                |
| DL           | 深度学习 (Deep Learning)                                         |
| DLC          | 深度学习容器 (Deep Learning Container)                             |
| DSP          | 数字信号处理器 (Digital Signal Processor)                           |
| GenAI        | 生成式 AI (Generative AI)                                       |
| GPU          | 图形处理单元 (Graphics Processing Unit)                            |
| HMX          | Hexagon 矩阵扩展 (Hexagon Matrix Extensions)                     |
| HTA          | Hexagon 张量加速器 (Hexagon Tensor Accelerator)                   |
| HTP          | Hexagon 张量处理器 (Hexagon Tensor Processor)                     |
| IOT          | 物联网 (Internet Of Things)                                     |
| Llama.cpp    | 一个 C/C++ 库和 CLI,用于在 CPU/GPU 上使用量化的 GGUF 模型本地运行 LLM 和 VLM     |
| LLM          | 大语言模型 (Large Language Model)                                 |
| ML           | 机器学习 (Machine Learning)                                      |
| NDK          | 原生开发工具包 (Native Development Kit)                             |
| NN           | 神经网络 (Neural Network)                                        |
| ONNX         | 开放神经网络交换 (Open Neural Network Exchange)                      |
| OpenCL       | 开放计算语言 (Open Computing Language)                             |
| PD           | 进程域 (Process Domain)                                         |
| PPA          | 个人软件包存档 (Personal Package Archive)                           |
| QNN          | Qualcomm 神经网络 (Qualcomm® AI Engine Direct 的同义词)              |
| QSML         | Qualcomm Snapdragon 数学库 (Qualcomm Snapdragon Math Libraries) |
| QTI          | Qualcomm Technologies Incorporated                           |
| RPC          | 远程过程调用 (Remote Procedure Call)                               |
| SDK          | 软件开发工具包 (Software Development Kit)                           |
| SNPE         | Snapdragon 神经处理引擎 (Qualcomm® Neural Processing SDK 的同义词)     |
| T2T          | 文本到文本 (Text-to-Text)                                         |
| TF           | Tensorflow                                                   |
| TTS          | 文本转语音 (Text-to-Speech)                                       |
| UDO          | 用户自定义操作 (User Defined Operations)                            |
| ViT          | Vision Transformer                                           |
| VLM          | 视觉语言模型 (Vision-Language Model)                               |
| Voice AI SDK | Qualcomm 的设备端自动语音识别引擎 (VoiceAI ASR),在 NPU 上运行                |
| VTCM         | 紧密耦合内存 (Very Tightly Coupled Memory)                         |
| XML          | 可扩展标记语言 (Extensible Markup Language)                         |
| WSL          | Windows Subsystem for Linux                                  |