diff --git a/backend/app/api/auto_download.py b/backend/app/api/auto_download.py
index 8f02b6c..6097547 100644
--- a/backend/app/api/auto_download.py
+++ b/backend/app/api/auto_download.py
@@ -3,6 +3,7 @@ from sqlalchemy.ext.asyncio import AsyncSession
 from sqlalchemy import select
 from typing import List, Optional
 import json
+import logging
 
 from app.db.session import get_db
 from app.models.models import DownloadJob, Music
@@ -14,6 +15,7 @@ import os
 from datetime import datetime
 
 router = APIRouter()
+logger = logging.getLogger(__name__)
 
 
 async def process_download_job(job_id: int, db: AsyncSession):
@@ -30,6 +32,8 @@ async def process_download_job(job_id: int, db: AsyncSession):
         await db.commit()
 
         # Search for the song
+        # If artist is included in song_name (format: "Artist - Song"), use it directly
+        # Otherwise, just search by song name
         search_results = await music_searcher.search_all(job.song_name, limit=10)
 
         # Combine results and filter out short clips (< 90 seconds) to avoid samples
@@ -86,6 +90,8 @@ async def process_download_job(job_id: int, db: AsyncSession):
         # Format output filename as "artist - title" to avoid downloading wrong songs
         output_filename = f"{selected_result.artist} - {selected_result.title}" if selected_result.artist else selected_result.title
         
+        logger.info(f"Auto-download job {job_id}: artist='{selected_result.artist}', title='{selected_result.title}', output_filename='{output_filename}'")
+        
         # Download the music
         success, file_path, error = await music_downloader.download_music(
             selected_result.url,
@@ -150,9 +156,18 @@ async def create_download_job(
     db: AsyncSession = Depends(get_db)
 ):
     """Create a new auto-download job"""
+    # Format song_name to include artist if provided
+    # This ensures better search results
+    song_name = request.song_name
+    if request.artist:
+        # Format as "Artist - Song" for better search accuracy
+        song_name = f"{request.artist} - {request.song_name}"
+    
+    logger.info(f"Creating auto-download job: song_name='{song_name}' (original: '{request.song_name}', artist: '{request.artist}')")
+    
     # Create job
     job = DownloadJob(
-        song_name=request.song_name,
+        song_name=song_name,
         status="pending"
     )
 
diff --git a/backend/app/api/download.py b/backend/app/api/download.py
index a962d0b..6d6c2e8 100644
--- a/backend/app/api/download.py
+++ b/backend/app/api/download.py
@@ -1,6 +1,7 @@
 from fastapi import APIRouter, Depends, HTTPException, BackgroundTasks
 from sqlalchemy.ext.asyncio import AsyncSession
 from sqlalchemy import select
+import logging
 
 from app.db.session import get_db
 from app.models.models import Music
@@ -13,6 +14,7 @@ import os
 import uuid
 
 router = APIRouter()
+logger = logging.getLogger(__name__)
 
 
 async def process_download(
@@ -29,6 +31,8 @@ async def process_download(
     # Format output filename as "artist - title" if artist is provided
     output_filename = f"{artist} - {title}" if artist else title
     
+    logger.info(f"Download task {task_id}: artist='{artist}', title='{title}', output_filename='{output_filename}'")
+    
     success, file_path, error = await music_downloader.download_music(url, output_filename)
     
     if success and file_path:
diff --git a/backend/app/schemas/schemas.py b/backend/app/schemas/schemas.py
index 3bb2fe8..03b3f16 100644
--- a/backend/app/schemas/schemas.py
+++ b/backend/app/schemas/schemas.py
@@ -147,6 +147,7 @@ class MusicDetailInfo(BaseModel):
 
 class DownloadJobCreate(BaseModel):
     song_name: str
+    artist: Optional[str] = None  # Artist name to improve search accuracy
 
 
 class DownloadJobResponse(BaseModel):
diff --git a/backend/app/services/search.py b/backend/app/services/search.py
index bdc4e56..be4ad46 100644
--- a/backend/app/services/search.py
+++ b/backend/app/services/search.py
@@ -3,10 +3,58 @@ import re
 from typing import List, Optional
 from app.schemas.schemas import SearchResult
 import logging
+import asyncio
 
 logger = logging.getLogger(__name__)
 
 
+def extract_artist_from_title(title: str) -> tuple[str, str]:
+    """
+    Try to extract artist and song name from YouTube video title.
+    Common patterns:
+    - "Artist - Song Title"
+    - "Artist: Song Title"
+    - "Song Title by Artist"
+    - "Artist | Song Title"
+    
+    Returns: (artist, clean_title)
+    """
+    # Pattern 1: "Artist - Song Title"
+    match = re.match(r'^([^-]+)\s*-\s*(.+)$', title)
+    if match:
+        artist = match.group(1).strip()
+        song = match.group(2).strip()
+        # Remove common suffixes like (Official Video), [Official Audio], etc.
+        song = re.sub(r'\s*[\(\[].*?(official|audio|video|lyric|mv).*?[\)\]]', '', song, flags=re.IGNORECASE)
+        return artist, song.strip()
+    
+    # Pattern 2: "Artist: Song Title"
+    match = re.match(r'^([^:]+):\s*(.+)$', title)
+    if match:
+        artist = match.group(1).strip()
+        song = match.group(2).strip()
+        song = re.sub(r'\s*[\(\[].*?(official|audio|video|lyric|mv).*?[\)\]]', '', song, flags=re.IGNORECASE)
+        return artist, song.strip()
+    
+    # Pattern 3: "Song Title by Artist"
+    match = re.search(r'^(.+?)\s+by\s+([^(\[]+)', title, re.IGNORECASE)
+    if match:
+        song = match.group(1).strip()
+        artist = match.group(2).strip()
+        return artist, song
+    
+    # Pattern 4: "Artist | Song Title"
+    match = re.match(r'^([^|]+)\|\s*(.+)$', title)
+    if match:
+        artist = match.group(1).strip()
+        song = match.group(2).strip()
+        song = re.sub(r'\s*[\(\[].*?(official|audio|video|lyric|mv).*?[\)\]]', '', song, flags=re.IGNORECASE)
+        return artist, song.strip()
+    
+    # No pattern matched, return empty artist and original title
+    return "", title
+
+
 class MusicSearcher:
     """Search for music from various sources (similar to xiaomusic search logic)"""
     
@@ -44,9 +92,29 @@ class MusicSearcher:
                     for line in lines:
                         try:
                             data = json.loads(line)
+                            title = data.get('title', '')
+                            
+                            # Try to get artist from metadata first
+                            artist = (
+                                data.get('artist') or 
+                                data.get('creator') or 
+                                ''
+                            )
+                            
+                            # If no artist in metadata, try to extract from title
+                            if not artist:
+                                extracted_artist, clean_title = extract_artist_from_title(title)
+                                if extracted_artist:
+                                    artist = extracted_artist
+                                    title = clean_title
+                            
+                            # If still no artist, use uploader as last resort
+                            if not artist:
+                                artist = data.get('uploader', '')
+                            
                             results.append(SearchResult(
-                                title=data.get('title', ''),
-                                artist=data.get('uploader', ''),
+                                title=title,
+                                artist=artist,
                                 duration=data.get('duration', 0),
                                 thumbnail=data.get('thumbnail', ''),
                                 url=data.get('webpage_url', ''),
@@ -94,9 +162,19 @@ class MusicSearcher:
                         
                         if data.get('code') == 0 and 'data' in data:
                             for item in data['data'].get('result', []):
+                                title = item.get('title', '').replace('<em class="keyword">', '').replace('</em>', '')
+                                artist = item.get('author', '')
+                                
+                                # If no artist, try to extract from title
+                                if not artist:
+                                    extracted_artist, clean_title = extract_artist_from_title(title)
+                                    if extracted_artist:
+                                        artist = extracted_artist
+                                        title = clean_title
+                                
                                 results.append(SearchResult(
-                                    title=item.get('title', '').replace('<em class="keyword">', '').replace('</em>', ''),
-                                    artist=item.get('author', ''),
+                                    title=title,
+                                    artist=artist,
                                     duration=item.get('duration', 0),
                                     thumbnail=f"https:{item.get('pic', '')}" if item.get('pic') else '',
                                     url=item.get('arcurl', ''),
diff --git a/frontend/src/api/client.ts b/frontend/src/api/client.ts
index 8b61810..0d4ff06 100644
--- a/frontend/src/api/client.ts
+++ b/frontend/src/api/client.ts
@@ -100,8 +100,8 @@ export const settingsApi = {
 
 // Auto Download API
 export const autoDownloadApi = {
-  createJob: (songName: string) => 
-    api.post('/auto-download/job', { song_name: songName }),
+  createJob: (songName: string, artist?: string) => 
+    api.post('/auto-download/job', { song_name: songName, artist }),
   getJobs: (status?: string) => 
     api.get('/auto-download/jobs', { params: { status } }),
   getJob: (jobId: number) => 
diff --git a/frontend/src/components/artist/ArtistDetailPage.tsx b/frontend/src/components/artist/ArtistDetailPage.tsx
index bd79ff4..547f3c5 100644
--- a/frontend/src/components/artist/ArtistDetailPage.tsx
+++ b/frontend/src/components/artist/ArtistDetailPage.tsx
@@ -68,8 +68,8 @@ export default function ArtistDetailPage({ onPlayMusic }: ArtistDetailProps) {
   })
 
   const autoDownloadMutation = useMutation({
-    mutationFn: async (songName: string) => {
-      const response = await autoDownloadApi.createJob(songName)
+    mutationFn: async ({ songTitle, artist }: { songTitle: string; artist: string }) => {
+      const response = await autoDownloadApi.createJob(songTitle, artist)
       return response.data
     },
     onSuccess: () => {
@@ -362,7 +362,10 @@ export default function ArtistDetailPage({ onPlayMusic }: ArtistDetailProps) {
               Cancel
             </Button>
             <Button
-              onClick={() => selectedSong && autoDownloadMutation.mutate(selectedSong.title)}
+              onClick={() => selectedSong && autoDownloadMutation.mutate({ 
+                songTitle: selectedSong.title, 
+                artist: selectedSong.artist 
+              })}
               disabled={autoDownloadMutation.isPending}
             >
               {autoDownloadMutation.isPending ? (