Implement avatar upload and management for voice profiles

- Added functionality to upload, delete, and retrieve avatar images for voice profiles.
- Introduced new API endpoints for avatar management, including upload and delete operations.
- Enhanced profile forms and components to support avatar image handling, including previews and error handling.
- Updated database schema to include avatar_path for profiles and added necessary migrations.
- Implemented image validation and processing utilities to ensure proper avatar uploads.

app/src/components/Generation/FloatingGenerateBox.tsx

@@ -112,7 +112,6 @@ export function FloatingGenerateBox({
     }
   }, [selectedProfileId, profiles, setSelectedProfileId]);
 
-
   // Auto-resize textarea based on content (only when expanded)
   useEffect(() => {
     if (!isExpanded) {

app/src/components/StoriesTab/StoryChatItem.tsx

@@ -1,6 +1,7 @@
 import { useSortable } from '@dnd-kit/sortable';
 import { CSS } from '@dnd-kit/utilities';
 import { GripVertical, Mic, MoreHorizontal, Play, Trash2 } from 'lucide-react';
+import { useState } from 'react';
 import { Button } from '@/components/ui/button';
 import {
   DropdownMenu,
@@ -12,6 +13,7 @@ import { Textarea } from '@/components/ui/textarea';
 import type { StoryItemDetail } from '@/lib/api/types';
 import { cn } from '@/lib/utils/cn';
 import { useStoryStore } from '@/stores/storyStore';
+import { useServerStore } from '@/stores/serverStore';
 
 interface StoryChatItemProps {
   item: StoryItemDetail;
@@ -33,6 +35,10 @@ export function StoryChatItem({
   isDragging,
 }: StoryChatItemProps) {
   const seek = useStoryStore((state) => state.seek);
+  const serverUrl = useServerStore((state) => state.serverUrl);
+  const [avatarError, setAvatarError] = useState(false);
+
+  const avatarUrl = `${serverUrl}/profiles/${item.profile_id}/avatar`;
 
   // Check if this item is currently playing based on timecode
   const itemStartMs = item.start_time_ms;
@@ -72,10 +78,22 @@ export function StoryChatItem({
         </button>
       )}
 
-      {/* Voice Icon */}
+      {/* Voice Avatar */}
       <div className="shrink-0">
-        <div className="h-10 w-10 rounded-full bg-muted flex items-center justify-center">
-          <Mic className="h-5 w-5 text-muted-foreground" />
+        <div className="h-10 w-10 rounded-full bg-muted flex items-center justify-center overflow-hidden">
+          {!avatarError ? (
+            <img
+              src={avatarUrl}
+              alt={`${item.profile_name} avatar`}
+              className={cn(
+                'h-full w-full object-cover transition-all duration-200',
+                !isCurrentlyPlaying && 'grayscale'
+              )}
+              onError={() => setAvatarError(true)}
+            />
+          ) : (
+            <Mic className="h-5 w-5 text-muted-foreground" />
+          )}
         </div>
       </div>
 

app/src/components/VoiceProfiles/ProfileCard.tsx

@@ -1,5 +1,6 @@
 import { Download, Edit, Mic, Trash2 } from 'lucide-react';
 import { useState } from 'react';
+import { useServerStore } from '@/stores/serverStore';
 import { Badge } from '@/components/ui/badge';
 import { Button } from '@/components/ui/button';
 import { Card, CardContent, CardHeader, CardTitle } from '@/components/ui/card';
@@ -23,15 +24,21 @@ interface ProfileCardProps {
 
 export function ProfileCard({ profile }: ProfileCardProps) {
   const [deleteDialogOpen, setDeleteDialogOpen] = useState(false);
+  const [avatarError, setAvatarError] = useState(false);
   const deleteProfile = useDeleteProfile();
   const exportProfile = useExportProfile();
   const setEditingProfileId = useUIStore((state) => state.setEditingProfileId);
   const setProfileDialogOpen = useUIStore((state) => state.setProfileDialogOpen);
   const selectedProfileId = useUIStore((state) => state.selectedProfileId);
   const setSelectedProfileId = useUIStore((state) => state.setSelectedProfileId);
+  const serverUrl = useServerStore((state) => state.serverUrl);
 
   const isSelected = selectedProfileId === profile.id;
 
+  const avatarUrl = profile.avatar_path
+    ? `${serverUrl}/profiles/${profile.id}/avatar`
+    : null;
+
   const handleSelect = () => {
     setSelectedProfileId(isSelected ? null : profile.id);
   };
@@ -67,8 +74,20 @@ export function ProfileCard({ profile }: ProfileCardProps) {
       >
         <CardHeader className="p-3 pb-2">
           <CardTitle className="flex items-center gap-1.5 text-base font-medium">
-            <div className="h-6 w-6 rounded-full bg-muted flex items-center justify-center shrink-0">
-              <Mic className="h-3.5 w-3.5 text-muted-foreground" />
+            <div className="h-6 w-6 rounded-full bg-muted flex items-center justify-center shrink-0 overflow-hidden">
+              {avatarUrl && !avatarError ? (
+                <img
+                  src={avatarUrl}
+                  alt={`${profile.name} avatar`}
+                  className={cn(
+                    'h-full w-full object-cover transition-all duration-200',
+                    !isSelected && 'grayscale'
+                  )}
+                  onError={() => setAvatarError(true)}
+                />
+              ) : (
+                <Mic className="h-3.5 w-3.5 text-muted-foreground" />
+              )}
             </div>
             <span className="break-words">{profile.name}</span>
           </CardTitle>

app/src/components/VoiceProfiles/ProfileForm.tsx

@@ -1,6 +1,6 @@
 import { zodResolver } from '@hookform/resolvers/zod';
-import { Mic, Monitor, Upload, X } from 'lucide-react';
-import { useEffect, useState } from 'react';
+import { Edit2, Mic, Monitor, Upload, X } from 'lucide-react';
+import { useEffect, useRef, useState } from 'react';
 import { useForm } from 'react-hook-form';
 import * as z from 'zod';
 import { Button } from '@/components/ui/button';
@@ -36,14 +36,17 @@ import { useAudioRecording } from '@/lib/hooks/useAudioRecording';
 import {
   useAddSample,
   useCreateProfile,
+  useDeleteAvatar,
   useProfile,
   useUpdateProfile,
+  useUploadAvatar,
 } from '@/lib/hooks/useProfiles';
 import { useSystemAudioCapture } from '@/lib/hooks/useSystemAudioCapture';
 import { useTranscription } from '@/lib/hooks/useTranscription';
 import { isTauri } from '@/lib/tauri';
 import { formatAudioDuration, getAudioDuration } from '@/lib/utils/audio';
 import { type ProfileFormDraft, useUIStore } from '@/stores/uiStore';
+import { useServerStore } from '@/stores/serverStore';
 import { AudioSampleRecording } from './AudioSampleRecording';
 import { AudioSampleSystem } from './AudioSampleSystem';
 import { AudioSampleUpload } from './AudioSampleUpload';
@@ -57,6 +60,7 @@ const baseProfileSchema = z.object({
   language: z.enum(LANGUAGE_CODES as [LanguageCode, ...LanguageCode[]]),
   sampleFile: z.instanceof(File).optional(),
   referenceText: z.string().max(1000).optional(),
+  avatarFile: z.instanceof(File).optional(),
 });
 
 const profileSchema = baseProfileSchema.refine(
@@ -108,13 +112,18 @@ export function ProfileForm() {
   const createProfile = useCreateProfile();
   const updateProfile = useUpdateProfile();
   const addSample = useAddSample();
+  const uploadAvatar = useUploadAvatar();
+  const deleteAvatar = useDeleteAvatar();
   const transcribe = useTranscription();
   const { toast } = useToast();
   const [sampleMode, setSampleMode] = useState<'upload' | 'record' | 'system'>('record');
   const [audioDuration, setAudioDuration] = useState<number | null>(null);
   const [isValidatingAudio, setIsValidatingAudio] = useState(false);
+  const [avatarPreview, setAvatarPreview] = useState<string | null>(null);
+  const avatarInputRef = useRef<HTMLInputElement>(null);
   const { isPlaying, playPause, cleanup: cleanupAudio } = useAudioPlayer();
   const isCreating = !editingProfileId;
+  const serverUrl = useServerStore((state) => state.serverUrl);
 
   const form = useForm<ProfileFormValues>({
     resolver: zodResolver(profileSchema),
@@ -124,10 +133,12 @@ export function ProfileForm() {
       language: 'en',
       sampleFile: undefined,
       referenceText: '',
+      avatarFile: undefined,
     },
   });
 
   const selectedFile = form.watch('sampleFile');
+  const selectedAvatarFile = form.watch('avatarFile');
 
   // Validate audio duration when file is selected
   useEffect(() => {
@@ -244,6 +255,19 @@ export function ProfileForm() {
     }
   }, [systemRecordingError, toast]);
 
+  // Handle avatar preview
+  useEffect(() => {
+    if (selectedAvatarFile instanceof File) {
+      const url = URL.createObjectURL(selectedAvatarFile);
+      setAvatarPreview(url);
+      return () => URL.revokeObjectURL(url);
+    } else if (editingProfile?.avatar_path) {
+      setAvatarPreview(`${serverUrl}/profiles/${editingProfile.id}/avatar`);
+    } else {
+      setAvatarPreview(null);
+    }
+  }, [selectedAvatarFile, editingProfile, serverUrl]);
+
   // Restore form state from draft or editing profile
   useEffect(() => {
     if (editingProfile) {
@@ -253,6 +277,7 @@ export function ProfileForm() {
         language: editingProfile.language as LanguageCode,
         sampleFile: undefined,
         referenceText: undefined,
+        avatarFile: undefined,
       });
     } else if (profileFormDraft && open) {
       // Restore from draft when opening in create mode
@@ -262,6 +287,7 @@ export function ProfileForm() {
         language: profileFormDraft.language as LanguageCode,
         referenceText: profileFormDraft.referenceText,
         sampleFile: undefined,
+        avatarFile: undefined,
       });
       setSampleMode(profileFormDraft.sampleMode);
       // Restore the file if we have it saved
@@ -285,8 +311,10 @@ export function ProfileForm() {
         language: 'en',
         sampleFile: undefined,
         referenceText: undefined,
+        avatarFile: undefined,
       });
       setSampleMode('record');
+      setAvatarPreview(null);
     }
   }, [editingProfile, profileFormDraft, open, form]);
 
@@ -330,6 +358,52 @@ export function ProfileForm() {
     playPause(file);
   }
 
+  function handleAvatarFileChange(e: React.ChangeEvent<HTMLInputElement>) {
+    const file = e.target.files?.[0];
+    if (file) {
+      if (!file.type.startsWith('image/')) {
+        toast({
+          title: 'Invalid file type',
+          description: 'Please select an image file (PNG, JPG, or WebP)',
+          variant: 'destructive',
+        });
+        return;
+      }
+      if (file.size > 5 * 1024 * 1024) {
+        toast({
+          title: 'File too large',
+          description: 'Image must be less than 5MB',
+          variant: 'destructive',
+        });
+        return;
+      }
+      form.setValue('avatarFile', file);
+    }
+  }
+
+  async function handleRemoveAvatar() {
+    if (editingProfileId && editingProfile?.avatar_path) {
+      try {
+        await deleteAvatar.mutateAsync(editingProfileId);
+        toast({
+          title: 'Avatar removed',
+          description: 'Avatar image has been removed successfully.',
+        });
+      } catch (error) {
+        toast({
+          title: 'Failed to remove avatar',
+          description: error instanceof Error ? error.message : 'Unknown error',
+          variant: 'destructive',
+        });
+      }
+    }
+    form.setValue('avatarFile', undefined);
+    setAvatarPreview(null);
+    if (avatarInputRef.current) {
+      avatarInputRef.current.value = '';
+    }
+  }
+
   async function onSubmit(data: ProfileFormValues) {
     try {
       if (editingProfileId) {
@@ -342,6 +416,23 @@ export function ProfileForm() {
             language: data.language,
           },
         });
+
+        // Handle avatar upload/update if file changed
+        if (data.avatarFile) {
+          try {
+            await uploadAvatar.mutateAsync({
+              profileId: editingProfileId,
+              file: data.avatarFile,
+            });
+          } catch (avatarError) {
+            toast({
+              title: 'Avatar upload failed',
+              description: avatarError instanceof Error ? avatarError.message : 'Failed to upload avatar',
+              variant: 'destructive',
+            });
+          }
+        }
+
         toast({
           title: 'Voice updated',
           description: `"${data.name}" has been updated successfully.`,
@@ -418,6 +509,23 @@ export function ProfileForm() {
             file: sampleFile,
             referenceText: referenceText,
           });
+
+          // Handle avatar upload if provided
+          if (data.avatarFile) {
+            try {
+              await uploadAvatar.mutateAsync({
+                profileId: profile.id,
+                file: data.avatarFile,
+              });
+            } catch (avatarError) {
+              toast({
+                title: 'Avatar upload failed',
+                description: avatarError instanceof Error ? avatarError.message : 'Failed to upload avatar',
+                variant: 'destructive',
+              });
+            }
+          }
+
           toast({
             title: 'Profile created',
             description: `"${data.name}" has been created with a sample.`,
@@ -670,6 +778,58 @@ export function ProfileForm() {
 
                 {/* Right column: Profile info */}
                 <div className="space-y-4">
+                  {/* Avatar Upload */}
+                  <FormField
+                    control={form.control}
+                    name="avatarFile"
+                    render={() => (
+                      <FormItem>
+                        <FormControl>
+                          <div className="flex justify-center pt-4 pb-2">
+                            <div className="relative group">
+                              <div className="h-24 w-24 rounded-full bg-muted flex items-center justify-center shrink-0 overflow-hidden border-2 border-border">
+                                {avatarPreview ? (
+                                  <img
+                                    src={avatarPreview}
+                                    alt="Avatar preview"
+                                    className="h-full w-full object-cover"
+                                  />
+                                ) : (
+                                  <Mic className="h-10 w-10 text-muted-foreground" />
+                                )}
+                              </div>
+                              <button
+                                type="button"
+                                onClick={() => avatarInputRef.current?.click()}
+                                className="absolute inset-0 rounded-full bg-accent/60 opacity-0 group-hover:opacity-100 transition-opacity flex items-center justify-center cursor-pointer"
+                              >
+                                <Edit2 className="h-6 w-6 text-accent-foreground" />
+                              </button>
+                              {(avatarPreview || editingProfile?.avatar_path) && (
+                                <button
+                                  type="button"
+                                  onClick={handleRemoveAvatar}
+                                  disabled={deleteAvatar.isPending}
+                                  className="absolute bottom-0 right-0 h-6 w-6 rounded-full bg-destructive text-destructive-foreground flex items-center justify-center hover:bg-destructive/90 transition-colors shadow-sm"
+                                >
+                                  <X className="h-3.5 w-3.5" />
+                                </button>
+                              )}
+                            </div>
+                            <input
+                              ref={avatarInputRef}
+                              type="file"
+                              accept="image/png,image/jpeg,image/webp"
+                              onChange={handleAvatarFileChange}
+                              className="hidden"
+                            />
+                          </div>
+                        </FormControl>
+                        <FormMessage />
+                      </FormItem>
+                    )}
+                  />
+
                   <FormField
                     control={form.control}
                     name="name"

app/src/components/VoiceProfiles/SampleList.tsx

@@ -188,15 +188,7 @@ export function SampleList({ profileId }: SampleListProps) {
   }
 
   return (
-    <div className="space-y-4">
-      <div className="flex items-center justify-between">
-        <h3 className="text-lg font-semibold">Audio Samples</h3>
-        <Button type="button" size="sm" onClick={() => setUploadOpen(true)}>
-          <Plus className="mr-2 h-4 w-4" />
-          Add Sample
-        </Button>
-      </div>
-
+    <div className="space-y-4 pt-4">
       {samples && samples.length === 0 ? (
         <div className="flex flex-col items-center justify-center py-8 text-center border border-dashed rounded-lg">
           <Volume2 className="h-8 w-8 text-muted-foreground/50 mb-2" />
@@ -304,6 +296,11 @@ export function SampleList({ profileId }: SampleListProps) {
         </div>
       )}
 
+      <Button type="button" variant="outline" className="w-full" onClick={() => setUploadOpen(true)}>
+        <Plus className="mr-2 h-4 w-4" />
+        Add Sample
+      </Button>
+
       <SampleUpload profileId={profileId} open={uploadOpen} onOpenChange={setUploadOpen} />
     </div>
   );

app/src/lib/api/client.ts

@@ -165,6 +165,32 @@ class ApiClient {
     return response.json();
   }
 
+  async uploadAvatar(profileId: string, file: File): Promise<VoiceProfileResponse> {
+    const url = `${this.getBaseUrl()}/profiles/${profileId}/avatar`;
+    const formData = new FormData();
+    formData.append('file', file);
+
+    const response = await fetch(url, {
+      method: 'POST',
+      body: formData,
+    });
+
+    if (!response.ok) {
+      const error = await response.json().catch(() => ({
+        detail: response.statusText,
+      }));
+      throw new Error(error.detail || `HTTP error! status: ${response.status}`);
+    }
+
+    return response.json();
+  }
+
+  async deleteAvatar(profileId: string): Promise<void> {
+    await this.request<void>(`/profiles/${profileId}/avatar`, {
+      method: 'DELETE',
+    });
+  }
+
   // Generation
   async generateSpeech(data: GenerationRequest): Promise<GenerationResponse> {
     return this.request<GenerationResponse>('/generate', {

app/src/lib/api/types.ts

@@ -12,6 +12,7 @@ export interface VoiceProfileResponse {
   name: string;
   description?: string;
   language: string;
+  avatar_path?: string;
   created_at: string;
   updated_at: string;
 }

app/src/lib/hooks/useProfiles.ts

@@ -185,3 +185,32 @@ export function useImportProfile() {
     },
   });
 }
+
+export function useUploadAvatar() {
+  const queryClient = useQueryClient();
+
+  return useMutation({
+    mutationFn: ({ profileId, file }: { profileId: string; file: File }) =>
+      apiClient.uploadAvatar(profileId, file),
+    onSuccess: (_, variables) => {
+      queryClient.invalidateQueries({ queryKey: ['profiles'] });
+      queryClient.invalidateQueries({
+        queryKey: ['profiles', variables.profileId],
+      });
+    },
+  });
+}
+
+export function useDeleteAvatar() {
+  const queryClient = useQueryClient();
+
+  return useMutation({
+    mutationFn: (profileId: string) => apiClient.deleteAvatar(profileId),
+    onSuccess: (_, profileId) => {
+      queryClient.invalidateQueries({ queryKey: ['profiles'] });
+      queryClient.invalidateQueries({
+        queryKey: ['profiles', profileId],
+      });
+    },
+  });
+}

backend/database.py

@@ -17,11 +17,12 @@ Base = declarative_base()
 class VoiceProfile(Base):
     """Voice profile database model."""
     __tablename__ = "profiles"
-    
+
     id = Column(String, primary_key=True, default=lambda: str(uuid.uuid4()))
     name = Column(String, unique=True, nullable=False)
     description = Column(Text)
     language = Column(String, default="en")
+    avatar_path = Column(String, nullable=True)
     created_at = Column(DateTime, default=datetime.utcnow)
     updated_at = Column(DateTime, default=datetime.utcnow, onupdate=datetime.utcnow)
 
@@ -277,6 +278,16 @@ def _run_migrations(engine):
             conn.commit()
             print("Added trim_end_ms column to story_items")
 
+    # Migration: Add avatar_path to profiles table
+    if 'profiles' in inspector.get_table_names():
+        columns = {col['name'] for col in inspector.get_columns('profiles')}
+        if 'avatar_path' not in columns:
+            print("Migrating profiles: adding avatar_path column")
+            with engine.connect() as conn:
+                conn.execute(text("ALTER TABLE profiles ADD COLUMN avatar_path VARCHAR"))
+                conn.commit()
+                print("Added avatar_path column to profiles")
+
 
 def get_db():
     """Get database session (generator for dependency injection)."""

backend/export_import.py

@@ -75,6 +75,16 @@ def export_profile_to_zip(profile_id: str, db: Session) -> bytes:
     zip_buffer = io.BytesIO()
     
     with zipfile.ZipFile(zip_buffer, 'w', zipfile.ZIP_DEFLATED) as zip_file:
+        # Check if profile has avatar
+        has_avatar = False
+        if profile.avatar_path:
+            avatar_path = Path(profile.avatar_path)
+            if avatar_path.exists():
+                has_avatar = True
+                # Add avatar to ZIP root with original extension
+                avatar_ext = avatar_path.suffix
+                zip_file.write(avatar_path, f"avatar{avatar_ext}")
+
         # Create manifest.json
         manifest = {
             "version": "1.0",
@@ -82,30 +92,31 @@ def export_profile_to_zip(profile_id: str, db: Session) -> bytes:
                 "name": profile.name,
                 "description": profile.description,
                 "language": profile.language,
-            }
+            },
+            "has_avatar": has_avatar,
         }
         zip_file.writestr("manifest.json", json.dumps(manifest, indent=2))
-        
+
         # Create samples.json mapping
         samples_data = {}
         profile_dir = _get_profiles_dir() / profile_id
-        
+
         for sample in samples:
             # Get filename from audio_path (should be {sample_id}.wav)
             audio_path = Path(sample.audio_path)
             filename = audio_path.name
-            
+
             # Read audio file
             if not audio_path.exists():
                 raise ValueError(f"Audio file not found: {audio_path}")
-            
+
             # Add to samples directory in ZIP
             zip_path = f"samples/{filename}"
             zip_file.write(audio_path, zip_path)
-            
+
             # Map filename to reference text
             samples_data[filename] = sample.reference_text
-        
+
         zip_file.writestr("samples.json", json.dumps(samples_data, indent=2))
     
     zip_buffer.seek(0)
@@ -168,11 +179,31 @@ async def import_profile_from_zip(file_bytes: bytes, db: Session) -> VoiceProfil
             )
             
             profile = await create_profile(profile_create, db)
-            
+
             # Extract and add samples
             profile_dir = _get_profiles_dir() / profile.id
             profile_dir.mkdir(parents=True, exist_ok=True)
-            
+
+            # Handle avatar if present
+            avatar_files = [f for f in namelist if f.startswith("avatar.")]
+            if avatar_files:
+                try:
+                    avatar_file = avatar_files[0]
+                    # Extract to temporary file
+                    import tempfile
+                    with tempfile.NamedTemporaryFile(suffix=Path(avatar_file).suffix, delete=False) as tmp:
+                        tmp.write(zip_file.read(avatar_file))
+                        tmp_path = tmp.name
+
+                    try:
+                        from .profiles import upload_avatar
+                        await upload_avatar(profile.id, tmp_path, db)
+                    finally:
+                        Path(tmp_path).unlink(missing_ok=True)
+                except Exception as e:
+                    # Avatar import is optional - continue even if it fails
+                    pass
+
             for filename, reference_text in samples_data.items():
                 # Validate filename
                 if not filename.endswith('.wav'):

backend/main.py

@@ -296,6 +296,61 @@ async def update_profile_sample(
     return sample
 
 
+@app.post("/profiles/{profile_id}/avatar", response_model=models.VoiceProfileResponse)
+async def upload_profile_avatar(
+    profile_id: str,
+    file: UploadFile = File(...),
+    db: Session = Depends(get_db),
+):
+    """Upload or update avatar image for a profile."""
+    # Save uploaded file to temp location
+    with tempfile.NamedTemporaryFile(delete=False, suffix=Path(file.filename).suffix) as tmp:
+        content = await file.read()
+        tmp.write(content)
+        tmp_path = tmp.name
+
+    try:
+        profile = await profiles.upload_avatar(profile_id, tmp_path, db)
+        return profile
+    except ValueError as e:
+        raise HTTPException(status_code=400, detail=str(e))
+    finally:
+        # Clean up temp file
+        Path(tmp_path).unlink(missing_ok=True)
+
+
+@app.get("/profiles/{profile_id}/avatar")
+async def get_profile_avatar(
+    profile_id: str,
+    db: Session = Depends(get_db),
+):
+    """Get avatar image for a profile."""
+    profile = await profiles.get_profile(profile_id, db)
+    if not profile:
+        raise HTTPException(status_code=404, detail="Profile not found")
+
+    if not profile.avatar_path:
+        raise HTTPException(status_code=404, detail="No avatar found for this profile")
+
+    avatar_path = Path(profile.avatar_path)
+    if not avatar_path.exists():
+        raise HTTPException(status_code=404, detail="Avatar file not found")
+
+    return FileResponse(avatar_path)
+
+
+@app.delete("/profiles/{profile_id}/avatar")
+async def delete_profile_avatar(
+    profile_id: str,
+    db: Session = Depends(get_db),
+):
+    """Delete avatar image for a profile."""
+    success = await profiles.delete_avatar(profile_id, db)
+    if not success:
+        raise HTTPException(status_code=404, detail="Profile not found or no avatar to delete")
+    return {"message": "Avatar deleted successfully"}
+
+
 @app.get("/profiles/{profile_id}/export")
 async def export_profile(
     profile_id: str,

backend/models.py

@@ -20,6 +20,7 @@ class VoiceProfileResponse(BaseModel):
     name: str
     description: Optional[str]
     language: str
+    avatar_path: Optional[str] = None
     created_at: datetime
     updated_at: datetime
 

backend/profiles.py

@@ -21,6 +21,7 @@ from .database import (
     ProfileSample as DBProfileSample,
 )
 from .utils.audio import validate_reference_audio, load_audio, save_audio
+from .utils.images import validate_image, process_avatar
 from .tts import get_tts_model
 from . import config
 
@@ -307,23 +308,23 @@ async def create_voice_prompt_for_profile(
 ) -> dict:
     """
     Create a combined voice prompt from all samples in a profile.
-    
+
     Args:
         profile_id: Profile ID
         db: Database session
         use_cache: Whether to use cached prompts
-        
+
     Returns:
         Voice prompt dictionary
     """
     # Get all samples for profile
     samples = db.query(DBProfileSample).filter_by(profile_id=profile_id).all()
-    
+
     if not samples:
         raise ValueError(f"No samples found for profile {profile_id}")
-    
+
     tts_model = get_tts_model()
-    
+
     if len(samples) == 1:
         # Single sample - use directly
         sample = samples[0]
@@ -337,19 +338,19 @@ async def create_voice_prompt_for_profile(
         # Multiple samples - combine them
         audio_paths = [s.audio_path for s in samples]
         reference_texts = [s.reference_text for s in samples]
-        
+
         # Combine audio
         combined_audio, combined_text = await tts_model.combine_voice_prompts(
             audio_paths,
             reference_texts,
         )
-        
+
         # Save combined audio temporarily
         import tempfile
         with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as tmp:
             save_audio(combined_audio, tmp.name, 24000)
             tmp_path = tmp.name
-        
+
         try:
             # Create prompt from combined audio
             voice_prompt, _ = await tts_model.create_voice_prompt(
@@ -361,3 +362,99 @@ async def create_voice_prompt_for_profile(
         finally:
             # Clean up temp file
             Path(tmp_path).unlink(missing_ok=True)
+
+
+async def upload_avatar(
+    profile_id: str,
+    image_path: str,
+    db: Session,
+) -> VoiceProfileResponse:
+    """
+    Upload and process avatar image for a profile.
+
+    Args:
+        profile_id: Profile ID
+        image_path: Path to uploaded image file
+        db: Database session
+
+    Returns:
+        Updated profile
+    """
+    # Validate profile exists
+    profile = db.query(DBVoiceProfile).filter_by(id=profile_id).first()
+    if not profile:
+        raise ValueError(f"Profile {profile_id} not found")
+
+    # Validate image
+    is_valid, error_msg = validate_image(image_path)
+    if not is_valid:
+        raise ValueError(error_msg)
+
+    # Delete existing avatar if present
+    if profile.avatar_path:
+        old_avatar = Path(profile.avatar_path)
+        if old_avatar.exists():
+            old_avatar.unlink()
+
+    # Determine file extension from uploaded file
+    from PIL import Image
+    with Image.open(image_path) as img:
+        # Normalize JPEG variants (MPO is multi-picture format from some cameras)
+        img_format = img.format
+        if img_format in ('MPO', 'JPG'):
+            img_format = 'JPEG'
+        
+        ext_map = {
+            'PNG': '.png',
+            'JPEG': '.jpg',
+            'WEBP': '.webp'
+        }
+        ext = ext_map.get(img_format, '.png')
+
+    # Save processed image to profile directory
+    profile_dir = _get_profiles_dir() / profile_id
+    profile_dir.mkdir(parents=True, exist_ok=True)
+    output_path = profile_dir / f"avatar{ext}"
+
+    process_avatar(image_path, str(output_path))
+
+    # Update database
+    profile.avatar_path = str(output_path)
+    profile.updated_at = datetime.utcnow()
+
+    db.commit()
+    db.refresh(profile)
+
+    return VoiceProfileResponse.model_validate(profile)
+
+
+async def delete_avatar(
+    profile_id: str,
+    db: Session,
+) -> bool:
+    """
+    Delete avatar image for a profile.
+
+    Args:
+        profile_id: Profile ID
+        db: Database session
+
+    Returns:
+        True if deleted, False if not found or no avatar
+    """
+    profile = db.query(DBVoiceProfile).filter_by(id=profile_id).first()
+    if not profile or not profile.avatar_path:
+        return False
+
+    # Delete avatar file
+    avatar_path = Path(profile.avatar_path)
+    if avatar_path.exists():
+        avatar_path.unlink()
+
+    # Update database
+    profile.avatar_path = None
+    profile.updated_at = datetime.utcnow()
+
+    db.commit()
+
+    return True

backend/requirements.txt

@@ -21,3 +21,4 @@ numpy>=1.24.0
 
 # Utilities
 python-multipart>=0.0.6
+Pillow>=10.0.0

backend/utils/images.py

@@ -0,0 +1,114 @@
+"""Image processing utilities for avatar uploads."""
+
+from pathlib import Path
+from typing import Optional, Tuple
+from PIL import Image
+
+# JPEG can be reported as 'JPEG' or 'MPO' (for multi-picture format from some cameras)
+ALLOWED_FORMATS = {'PNG', 'JPEG', 'WEBP', 'MPO', 'JPG'}
+MAX_SIZE = 512
+MAX_FILE_SIZE = 5 * 1024 * 1024  # 5MB
+
+
+def validate_image(file_path: str) -> Tuple[bool, Optional[str]]:
+    """
+    Validate image format and file size.
+    
+    Args:
+        file_path: Path to image file
+        
+    Returns:
+        Tuple of (is_valid, error_message)
+    """
+    path = Path(file_path)
+    
+    # Check file size
+    if path.stat().st_size > MAX_FILE_SIZE:
+        return False, f"File size exceeds maximum of {MAX_FILE_SIZE // (1024 * 1024)}MB"
+    
+    try:
+        with Image.open(file_path) as img:
+            # Verify the image can be loaded
+            img.load()
+            
+            # Check format (normalize JPEG variants)
+            img_format = img.format
+            if img_format in ('MPO', 'JPG'):
+                img_format = 'JPEG'
+            
+            if img_format not in {'PNG', 'JPEG', 'WEBP'}:
+                return False, f"Invalid format '{img_format}'. Allowed formats: PNG, JPEG, WEBP"
+            
+            return True, None
+    except Exception as e:
+        return False, f"Invalid image file: {str(e)}"
+
+
+def process_avatar(input_path: str, output_path: str, max_size: int = MAX_SIZE) -> None:
+    """
+    Process avatar image: resize and optimize.
+    
+    Resizes image to fit within max_size x max_size while maintaining aspect ratio.
+    
+    Args:
+        input_path: Path to input image
+        output_path: Path to save processed image
+        max_size: Maximum width or height in pixels
+    """
+    with Image.open(input_path) as img:
+        # Handle EXIF orientation for JPEG images
+        try:
+            from PIL import ExifTags
+            for orientation in ExifTags.TAGS.keys():
+                if ExifTags.TAGS[orientation] == 'Orientation':
+                    break
+            exif = img._getexif()
+            if exif is not None:
+                orientation_value = exif.get(orientation)
+                if orientation_value == 3:
+                    img = img.rotate(180, expand=True)
+                elif orientation_value == 6:
+                    img = img.rotate(270, expand=True)
+                elif orientation_value == 8:
+                    img = img.rotate(90, expand=True)
+        except (AttributeError, KeyError, IndexError, TypeError):
+            # No EXIF data or orientation tag
+            pass
+        
+        # Convert to RGB if necessary (handles RGBA, P, CMYK, etc.)
+        if img.mode not in ('RGB', 'L'):
+            if img.mode == 'RGBA':
+                # Create white background for RGBA images
+                background = Image.new('RGB', img.size, (255, 255, 255))
+                background.paste(img, mask=img.split()[3])  # Use alpha channel as mask
+                img = background
+            elif img.mode == 'CMYK':
+                # Convert CMYK to RGB
+                img = img.convert('RGB')
+            elif img.mode == 'P':
+                # Convert palette mode to RGB
+                img = img.convert('RGB')
+            else:
+                img = img.convert('RGB')
+        
+        # Calculate new size maintaining aspect ratio
+        img.thumbnail((max_size, max_size), Image.Resampling.LANCZOS)
+        
+        # Determine output format from extension
+        output_ext = Path(output_path).suffix.lower()
+        
+        format_map = {
+            '.png': 'PNG',
+            '.jpeg': 'JPEG',
+            '.jpg': 'JPEG',
+            '.webp': 'WEBP'
+        }
+        
+        output_format = format_map.get(output_ext, 'PNG')
+        
+        # Save with optimization
+        save_kwargs = {'optimize': True}
+        if output_format == 'JPEG':
+            save_kwargs['quality'] = 90
+        
+        img.save(output_path, format=output_format, **save_kwargs)