enabled gpu

.devcontainer/devcontainer.json

@@ -1,26 +1,32 @@
-// For format details, see https://aka.ms/devcontainer.json. For config options, see the
-// README at: https://github.com/devcontainers/templates/tree/main/src/docker-existing-dockerfile
 {
-	"name": "Existing Dockerfile",
-	"build": {
-		// Sets the run context to one level up instead of the .devcontainer folder.
-		"context": "..",
-		// Update the 'dockerFile' property if you aren't using the standard 'Dockerfile' filename.
-		"dockerfile": "../Dockerfile"
-	}
+    "name": "FYP Backend (GPU)",
+    "image": "e226274b3239", // 直接使用您已有的鏡像 ID
 
-	// Features to add to the dev container. More info: https://containers.dev/features.
-	// "features": {},
+    // 這是最最最關鍵的部分！
+    "runArgs": [
+        "--gpus=all"
+    ],
 
-	// Use 'forwardPorts' to make a list of ports inside the container available locally.
-	// "forwardPorts": [],
+    // 轉發端口，以便您可以訪問 FastAPI
+    "forwardPorts": [8000],
 
-	// Uncomment the next line to run commands after the container is created.
-	// "postCreateCommand": "cat /etc/os-release",
+    // 將工作區掛載到容器中
+    "workspaceMount": "source=${localWorkspaceFolder},target=/workspaces/FYP-Backend,type=bind,consistency=cached",
+    "workspaceFolder": "/workspaces/FYP-Backend",
 
-	// Configure tool-specific properties.
-	// "customizations": {},
+    // 讓容器在 VS Code 關閉後保持運行
+    "shutdownAction": "none",
 
-	// Uncomment to connect as an existing user other than the container default. More info: https://aka.ms/dev-containers-non-root.
-	// "remoteUser": "devcontainer"
+    // 在容器創建後運行的命令 (可選，但推薦)
+    "postCreateCommand": "pip install -r requirements.txt",
+
+    // VS Code 擴展推薦 (可選)
+    "customizations": {
+        "vscode": {
+            "extensions": [
+                "ms-python.python",
+                "ms-python.vscode-pylance"
+            ]
+        }
+    }
 }

analyzer/ASR_en_us.py

@@ -7,6 +7,10 @@ from phonemizer import phonemize
 import numpy as np
 from datetime import datetime, timezone
 
+# 【【【【【 新增程式碼 #1：自動檢測可用設備 】】】】】
+DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
+print(f"INFO: ASR_fr_fr.py is configured to use device: {DEVICE}")
+
 # --- 1. 全域設定與模型載入函數 (保持不變) ---
 MODEL_NAME = "MultiBridge/wav2vec-LnNor-IPA-ft"
 MODEL_SAVE_PATH = "./ASRs/MultiBridge-wav2vec-LnNor-IPA-ft-local"
@@ -38,6 +42,7 @@ def load_model():
         
         processor = Wav2Vec2Processor.from_pretrained(MODEL_SAVE_PATH)
         model = Wav2Vec2ForCTC.from_pretrained(MODEL_SAVE_PATH)
+        model.to(DEVICE)  # 將模型移動到檢測到的設備上
         print("英文 (en-us) 模型和處理器載入成功！")
         return True
     except Exception as e:
@@ -95,6 +100,7 @@ def analyze(audio_file_path: str, target_sentence: str) -> dict:
         raise IOError(f"讀取或處理音訊時發生錯誤: {e}")
     
     input_values = processor(speech, sampling_rate=16000, return_tensors="pt").input_values
+    input_values = input_values.to(DEVICE)
     with torch.no_grad():
         logits = model(input_values).logits
     predicted_ids = torch.argmax(logits, dim=-1)

analyzer/ASR_fr_fr.py

@@ -10,6 +10,10 @@ import unicodedata
 import re
 import epitran
 
+# 【【【【【 新增程式碼 #1：自動檢測可用設備 】】】】】
+DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
+print(f"INFO: ASR_fr_fr.py is configured to use device: {DEVICE}")
+
 # --- 1. 全域設定與模型載入函數 (已修改為法語模型) ---
 MODEL_NAME = "Cnam-LMSSC/wav2vec2-french-phonemizer"
 MODEL_SAVE_PATH = "./ASRs/Cnam-LMSSC-wav2vec2-french-phonemizer-local"
@@ -41,6 +45,7 @@ def load_model():
         
         processor = Wav2Vec2Processor.from_pretrained(MODEL_SAVE_PATH)
         model = Wav2Vec2ForCTC.from_pretrained(MODEL_SAVE_PATH)
+        model.to(DEVICE)  # 將模型移動到檢測到的設備上
         print("法語 (fr-fr) 模型和處理器載入成功！")
         return True
     except Exception as e:
@@ -106,6 +111,7 @@ def analyze(audio_file_path: str, target_sentence: str) -> dict:
         raise IOError(f"讀取或處理音訊時發生錯誤: {e}")
     
     input_values = processor(speech, sampling_rate=16000, return_tensors="pt").input_values
+    input_values = input_values.to(DEVICE)
     with torch.no_grad():
         logits = model(input_values).logits
     predicted_ids = torch.argmax(logits, dim=-1)