From d4f32673ab4c034a3080f56cbc5eeed78fe2944e Mon Sep 17 00:00:00 2001
From: Soulter <905617992@qq.com>
Date: Fri, 10 Jan 2025 22:08:43 +0800
Subject: [PATCH] =?UTF-8?q?fix:=20=E4=BF=AE=E5=A4=8D=E6=8C=81=E4=B9=85?=
 =?UTF-8?q?=E5=8C=96=E9=97=AE=E9=A2=98?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 astrbot/dashboard/routes/chat.py |  5 +++--
 dashboard/src/views/ChatPage.vue |  9 ++++++---
 stt.py                           | 33 ++++++++++++++++++++++++++++++++
 3 files changed, 42 insertions(+), 5 deletions(-)
 create mode 100644 stt.py

diff --git a/astrbot/dashboard/routes/chat.py b/astrbot/dashboard/routes/chat.py
index 2e44db351..735673976 100644
--- a/astrbot/dashboard/routes/chat.py
+++ b/astrbot/dashboard/routes/chat.py
@@ -68,8 +68,9 @@ class ChatRoute(Route):
             
             conversation = self.db.get_webchat_conversation_by_user_id(username, conversation_id)
             try:
-                history = json.loads(conversation['history'])
-            except BaseException:
+                history = json.loads(conversation.history)
+            except BaseException as e:
+                print(e)
                 history = []
             history.append({
                 'type': 'user',
diff --git a/dashboard/src/views/ChatPage.vue b/dashboard/src/views/ChatPage.vue
index 0a60e4c23..fcf7f92a8 100644
--- a/dashboard/src/views/ChatPage.vue
+++ b/dashboard/src/views/ChatPage.vue
@@ -14,12 +14,13 @@ marked.setOptions({
     <v-card style="margin-bottom: 16px; width: 100%; background-color: #fff; height: 100%;">
         <v-card-text style="width: 100%; height: calc(100vh - 120px);">
             <div style="height: 100%; display: flex; gap: 16px;">
-                <div style="max-width: 120px;">
+                <div style="max-width: 200px;">
                     <!-- conversation -->
-                    <v-btn style="margin-bottom: 16px;" @click="newC">创建对话</v-btn>
-                    <v-btn style="margin-bottom: 16px;" v-if="currCid" @click="deleteConversation(currCid)" color="error">删除对话</v-btn>
+                    <v-btn variant="tonal" rounded="xl" style="margin-bottom: 16px; min-width: 200px;" @click="newC" :disabled="!currCid">+ 创建对话</v-btn>
+                    
                     <v-card class="mx-auto" min-width="200">
                         <v-list dense nav
+                        rounded="xl"
                         v-if="conversations.length > 0"
                         @update:selected="getConversationMessages"
                         >
@@ -36,6 +37,8 @@ marked.setOptions({
                         </v-list-item>
                         </v-list>
                     </v-card>
+
+                    <v-btn variant="tonal" rounded="xl" style="position: fixed; bottom: 48px; margin-bottom: 16px; min-width: 200px;" v-if="currCid" @click="deleteConversation(currCid)" color="error">删除此对话</v-btn>
                 </div>
                 <div style="height: 100%; width: 100%;">
                     <div style="height: calc(100% - 64px); overflow-y: auto; padding: 16px; " ref="messageContainer">
diff --git a/stt.py b/stt.py
new file mode 100644
index 000000000..9eee69b8a
--- /dev/null
+++ b/stt.py
@@ -0,0 +1,33 @@
+import torch
+from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
+from datasets import load_dataset
+
+
+device = "cuda:0" if torch.cuda.is_available() else "cpu"
+torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
+
+model_id = "openai/whisper-large-v3"
+
+model = AutoModelForSpeechSeq2Seq.from_pretrained(
+    model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True
+)
+model.to(device)
+
+processor = AutoProcessor.from_pretrained(model_id)
+
+pipe = pipeline(
+    "automatic-speech-recognition",
+    model=model,
+    tokenizer=processor.tokenizer,
+    feature_extractor=processor.feature_extractor,
+    chunk_length_s=30,
+    batch_size=16,  # batch size for inference - set based on your device
+    torch_dtype=torch_dtype,
+    device=device,
+)
+
+dataset = load_dataset("distil-whisper/librispeech_long", "clean", split="validation")
+sample = dataset[0]["audio"]
+
+result = pipe(sample)
+print(result["text"])