From 9e237e63d4069ad2d2e0ee4e544fb92a93ca93bc Mon Sep 17 00:00:00 2001
From: aki <arisu@ieee.org>
Date: Sun, 19 Apr 2026 00:19:49 +0900
Subject: [PATCH] feat: add icon support for speech providers and update
 related configurations

---
 .../web/src/components/provider-icon/icons.ts |  4 ++++
 .../speech/components/provider-setting.vue    | 22 +++++++++++++++++--
 apps/web/src/pages/speech/index.vue           | 18 +++++++++++++--
 conf/providers/alibabacloud-speech.yaml       |  8 +++++++
 conf/providers/deepgram-speech.yaml           |  7 ++++++
 conf/providers/edge.yaml                      |  2 +-
 conf/providers/elevenlabs-speech.yaml         |  7 ++++++
 conf/providers/microsoft-speech.yaml          |  8 +++++++
 conf/providers/minimax-speech.yaml            |  8 +++++++
 conf/providers/openai-speech.yaml             |  8 +++++++
 conf/providers/openrouter-speech.yaml         |  8 +++++++
 conf/providers/volcengine-speech.yaml         |  8 +++++++
 internal/tts/bootstrap.go                     |  6 ++++-
 internal/tts/registry.go                      |  8 +++++++
 internal/tts/service.go                       |  5 +++++
 internal/tts/types.go                         |  1 +
 packages/sdk/src/types.gen.ts                 |  1 +
 spec/docs.go                                  |  3 +++
 spec/swagger.json                             |  3 +++
 spec/swagger.yaml                             |  2 ++
 20 files changed, 131 insertions(+), 6 deletions(-)
 create mode 100644 conf/providers/alibabacloud-speech.yaml
 create mode 100644 conf/providers/deepgram-speech.yaml
 create mode 100644 conf/providers/elevenlabs-speech.yaml
 create mode 100644 conf/providers/microsoft-speech.yaml
 create mode 100644 conf/providers/minimax-speech.yaml
 create mode 100644 conf/providers/openai-speech.yaml
 create mode 100644 conf/providers/openrouter-speech.yaml
 create mode 100644 conf/providers/volcengine-speech.yaml

diff --git a/apps/web/src/components/provider-icon/icons.ts b/apps/web/src/components/provider-icon/icons.ts
index 6c76d4a0..ac71fb4d 100644
--- a/apps/web/src/components/provider-icon/icons.ts
+++ b/apps/web/src/components/provider-icon/icons.ts
@@ -35,6 +35,8 @@ import {
   Lmstudio,
   Meta,
   MetaColor,
+  Microsoft,
+  MicrosoftColor,
   Minimax,
   MinimaxColor,
   Mistral,
@@ -105,6 +107,8 @@ export const iconMap: Record<string, Component> = {
   'cohere-color': CohereColor,
   'azure': Azure,
   'azure-color': AzureColor,
+  'microsoft': Microsoft,
+  'microsoft-color': MicrosoftColor,
   'nvidia': Nvidia,
   'nvidia-color': NvidiaColor,
   'fireworks': Fireworks,
diff --git a/apps/web/src/pages/speech/components/provider-setting.vue b/apps/web/src/pages/speech/components/provider-setting.vue
index 6dfa5a57..23b475ec 100644
--- a/apps/web/src/pages/speech/components/provider-setting.vue
+++ b/apps/web/src/pages/speech/components/provider-setting.vue
@@ -1,7 +1,19 @@
 <template>
   <div class="p-4">
     <section class="flex items-center gap-3">
-      <Volume2 class="size-5" />
+      <span class="flex size-10 shrink-0 items-center justify-center rounded-full bg-muted">
+        <ProviderIcon
+          v-if="curProvider?.icon"
+          :icon="curProvider.icon"
+          size="1.5em"
+        />
+        <span
+          v-else
+          class="text-xs font-medium text-muted-foreground"
+        >
+          {{ getInitials(curProvider?.name) }}
+        </span>
+      </span>
       <div class="min-w-0">
         <h2 class="text-sm font-semibold truncate">
           {{ curProvider?.name }}
@@ -191,7 +203,7 @@ import {
   Switch,
 } from '@memohai/ui'
 import ModelConfigEditor from './model-config-editor.vue'
-import { ChevronDown, ChevronUp, Eye, EyeOff, Volume2 } from 'lucide-vue-next'
+import { ChevronDown, ChevronUp, Eye, EyeOff } from 'lucide-vue-next'
 import { computed, inject, reactive, ref, watch } from 'vue'
 import { toast } from 'vue-sonner'
 import { useI18n } from 'vue-i18n'
@@ -199,6 +211,7 @@ import { useQuery, useQueryCache } from '@pinia/colada'
 import { getSpeechModels, getSpeechProvidersMeta, putModelsById, putProvidersById } from '@memohai/sdk'
 import type { TtsSpeechModelResponse, TtsSpeechProviderResponse } from '@memohai/sdk'
 import LoadingButton from '@/components/loading-button/index.vue'
+import ProviderIcon from '@/components/provider-icon/index.vue'
 
 interface SpeechFieldSchema {
   key: string
@@ -233,6 +246,11 @@ interface SpeechProviderMeta {
   models?: SpeechModelMeta[]
 }
 
+function getInitials(name: string | undefined) {
+  const label = name?.trim() ?? ''
+  return label ? label.slice(0, 2).toUpperCase() : '?'
+}
+
 const { t } = useI18n()
 const curProvider = inject('curTtsProvider', ref<TtsSpeechProviderResponse>())
 const curProviderId = computed(() => curProvider.value?.id)
diff --git a/apps/web/src/pages/speech/index.vue b/apps/web/src/pages/speech/index.vue
index 766cc821..210984f6 100644
--- a/apps/web/src/pages/speech/index.vue
+++ b/apps/web/src/pages/speech/index.vue
@@ -18,6 +18,12 @@ import type { TtsSpeechProviderResponse } from '@memohai/sdk'
 import ProviderSetting from './components/provider-setting.vue'
 import { Volume2 } from 'lucide-vue-next'
 import MasterDetailSidebarLayout from '@/components/master-detail-sidebar-layout/index.vue'
+import ProviderIcon from '@/components/provider-icon/index.vue'
+
+function getInitials(name: string | undefined) {
+  const label = name?.trim() ?? ''
+  return label ? label.slice(0, 2).toUpperCase() : '?'
+}
 
 const { data: providerData } = useQuery({
   key: () => ['speech-providers'],
@@ -79,9 +85,17 @@ watch(filteredProviders, (list) => {
             >
               <span class="relative shrink-0">
                 <span class="flex size-7 items-center justify-center rounded-full bg-muted">
-                  <Volume2
-                    class="size-3.5 text-muted-foreground"
+                  <ProviderIcon
+                    v-if="item.icon"
+                    :icon="item.icon"
+                    size="1.25em"
                   />
+                  <span
+                    v-else
+                    class="text-xs font-medium text-muted-foreground"
+                  >
+                    {{ getInitials(item.name) }}
+                  </span>
                 </span>
                 <span
                   v-if="item.enable !== false"
diff --git a/conf/providers/alibabacloud-speech.yaml b/conf/providers/alibabacloud-speech.yaml
new file mode 100644
index 00000000..bd080393
--- /dev/null
+++ b/conf/providers/alibabacloud-speech.yaml
@@ -0,0 +1,8 @@
+name: Alibaba Cloud Speech
+client_type: alibabacloud-speech
+icon: bailian-color
+
+models:
+  - model_id: cosyvoice-tts
+    name: CosyVoice TTS
+    type: speech
diff --git a/conf/providers/deepgram-speech.yaml b/conf/providers/deepgram-speech.yaml
new file mode 100644
index 00000000..c7ec348a
--- /dev/null
+++ b/conf/providers/deepgram-speech.yaml
@@ -0,0 +1,7 @@
+name: Deepgram Speech
+client_type: deepgram-speech
+
+models:
+  - model_id: deepgram-tts
+    name: Deepgram TTS
+    type: speech
diff --git a/conf/providers/edge.yaml b/conf/providers/edge.yaml
index 7d2e267a..c7d73f32 100644
--- a/conf/providers/edge.yaml
+++ b/conf/providers/edge.yaml
@@ -1,6 +1,6 @@
 name: Edge
 client_type: edge-speech
-icon: edge
+icon: microsoft
 
 models:
   - model_id: edge-read-aloud
diff --git a/conf/providers/elevenlabs-speech.yaml b/conf/providers/elevenlabs-speech.yaml
new file mode 100644
index 00000000..16e4d92d
--- /dev/null
+++ b/conf/providers/elevenlabs-speech.yaml
@@ -0,0 +1,7 @@
+name: ElevenLabs Speech
+client_type: elevenlabs-speech
+
+models:
+  - model_id: elevenlabs-tts
+    name: ElevenLabs TTS
+    type: speech
diff --git a/conf/providers/microsoft-speech.yaml b/conf/providers/microsoft-speech.yaml
new file mode 100644
index 00000000..2e79f25c
--- /dev/null
+++ b/conf/providers/microsoft-speech.yaml
@@ -0,0 +1,8 @@
+name: Microsoft Speech
+client_type: microsoft-speech
+icon: azure-color
+
+models:
+  - model_id: microsoft-tts
+    name: Microsoft TTS
+    type: speech
diff --git a/conf/providers/minimax-speech.yaml b/conf/providers/minimax-speech.yaml
new file mode 100644
index 00000000..b5acffeb
--- /dev/null
+++ b/conf/providers/minimax-speech.yaml
@@ -0,0 +1,8 @@
+name: MiniMax Speech
+client_type: minimax-speech
+icon: minimax-color
+
+models:
+  - model_id: minimax-tts
+    name: MiniMax TTS
+    type: speech
diff --git a/conf/providers/openai-speech.yaml b/conf/providers/openai-speech.yaml
new file mode 100644
index 00000000..6364c716
--- /dev/null
+++ b/conf/providers/openai-speech.yaml
@@ -0,0 +1,8 @@
+name: OpenAI Speech
+client_type: openai-speech
+icon: openai
+
+models:
+  - model_id: gpt-4o-mini-tts
+    name: GPT-4o Mini TTS
+    type: speech
diff --git a/conf/providers/openrouter-speech.yaml b/conf/providers/openrouter-speech.yaml
new file mode 100644
index 00000000..c31c0c3e
--- /dev/null
+++ b/conf/providers/openrouter-speech.yaml
@@ -0,0 +1,8 @@
+name: OpenRouter Speech
+client_type: openrouter-speech
+icon: openrouter
+
+models:
+  - model_id: openrouter-tts
+    name: OpenRouter TTS
+    type: speech
diff --git a/conf/providers/volcengine-speech.yaml b/conf/providers/volcengine-speech.yaml
new file mode 100644
index 00000000..126226ce
--- /dev/null
+++ b/conf/providers/volcengine-speech.yaml
@@ -0,0 +1,8 @@
+name: Volcengine Speech
+client_type: volcengine-speech
+icon: volcengine-color
+
+models:
+  - model_id: sami-tts
+    name: SAMI TTS
+    type: speech
diff --git a/internal/tts/bootstrap.go b/internal/tts/bootstrap.go
index 19abaa52..4196a61d 100644
--- a/internal/tts/bootstrap.go
+++ b/internal/tts/bootstrap.go
@@ -17,11 +17,15 @@ func SyncRegistry(ctx context.Context, logger *slog.Logger, queries *sqlc.Querie
 		if err != nil {
 			return fmt.Errorf("marshal speech provider config: %w", err)
 		}
+		var icon pgtype.Text
+		if def.Icon != "" {
+			icon = pgtype.Text{String: def.Icon, Valid: true}
+		}
 
 		provider, err := queries.UpsertRegistryProvider(ctx, sqlc.UpsertRegistryProviderParams{
 			Name:       def.DisplayName,
 			ClientType: string(def.ClientType),
-			Icon:       pgtype.Text{},
+			Icon:       icon,
 			Config:     configJSON,
 		})
 		if err != nil {
diff --git a/internal/tts/registry.go b/internal/tts/registry.go
index 7e4b8c79..25348e6c 100644
--- a/internal/tts/registry.go
+++ b/internal/tts/registry.go
@@ -25,6 +25,7 @@ type ProviderFactory func(config map[string]any) (sdk.SpeechProvider, error)
 type ProviderDefinition struct {
 	ClientType   models.ClientType
 	DisplayName  string
+	Icon         string
 	Description  string
 	ConfigSchema ConfigSchema
 	DefaultModel string
@@ -123,6 +124,7 @@ func defaultProviderDefinitions() []ProviderDefinition {
 		{
 			ClientType:   models.ClientTypeEdgeSpeech,
 			DisplayName:  "Microsoft Edge",
+			Icon:         "microsoft",
 			Description:  "Free Edge Read Aloud TTS",
 			ConfigSchema: ConfigSchema{Fields: []FieldSchema{stringField("base_url", "Base URL", "Override the Edge WebSocket endpoint", false, "", 10)}},
 			DefaultModel: "edge-read-aloud",
@@ -163,6 +165,7 @@ func defaultProviderDefinitions() []ProviderDefinition {
 		{
 			ClientType:  models.ClientTypeOpenAISpeech,
 			DisplayName: "OpenAI Speech",
+			Icon:        "openai",
 			Description: "OpenAI /audio/speech compatible TTS",
 			ConfigSchema: ConfigSchema{Fields: []FieldSchema{
 				secretField("api_key", "API Key", "Bearer API key", true, 10),
@@ -204,6 +207,7 @@ func defaultProviderDefinitions() []ProviderDefinition {
 		{
 			ClientType:  models.ClientTypeOpenRouterSpeech,
 			DisplayName: "OpenRouter Speech",
+			Icon:        "openrouter",
 			Description: "OpenRouter audio modality TTS",
 			ConfigSchema: ConfigSchema{Fields: []FieldSchema{
 				secretField("api_key", "API Key", "OpenRouter API key", true, 10),
@@ -333,6 +337,7 @@ func defaultProviderDefinitions() []ProviderDefinition {
 		{
 			ClientType:  models.ClientTypeMiniMaxSpeech,
 			DisplayName: "MiniMax Speech",
+			Icon:        "minimax-color",
 			Description: "MiniMax TTS",
 			ConfigSchema: ConfigSchema{Fields: []FieldSchema{
 				secretField("api_key", "API Key", "MiniMax API key", true, 10),
@@ -380,6 +385,7 @@ func defaultProviderDefinitions() []ProviderDefinition {
 		{
 			ClientType:  models.ClientTypeVolcengineSpeech,
 			DisplayName: "Volcengine Speech",
+			Icon:        "volcengine-color",
 			Description: "Volcengine SAMI TTS",
 			ConfigSchema: ConfigSchema{Fields: []FieldSchema{
 				secretField("access_key", "Access Key", "Volcengine access key", true, 10),
@@ -431,6 +437,7 @@ func defaultProviderDefinitions() []ProviderDefinition {
 		{
 			ClientType:  models.ClientTypeAlibabaSpeech,
 			DisplayName: "Alibaba Cloud Speech",
+			Icon:        "bailian-color",
 			Description: "DashScope CosyVoice TTS",
 			ConfigSchema: ConfigSchema{Fields: []FieldSchema{
 				secretField("api_key", "API Key", "DashScope API key", true, 10),
@@ -478,6 +485,7 @@ func defaultProviderDefinitions() []ProviderDefinition {
 		{
 			ClientType:  models.ClientTypeMicrosoftSpeech,
 			DisplayName: "Microsoft Speech",
+			Icon:        "azure-color",
 			Description: "Azure Cognitive Services TTS",
 			ConfigSchema: ConfigSchema{Fields: []FieldSchema{
 				secretField("api_key", "API Key", "Azure speech subscription key", true, 10),
diff --git a/internal/tts/service.go b/internal/tts/service.go
index 816b55a0..2734e55b 100644
--- a/internal/tts/service.go
+++ b/internal/tts/service.go
@@ -209,10 +209,15 @@ func mergeConfig(parts ...map[string]any) map[string]any {
 }
 
 func toSpeechProviderResponse(row sqlc.Provider) SpeechProviderResponse {
+	icon := ""
+	if row.Icon.Valid {
+		icon = row.Icon.String
+	}
 	return SpeechProviderResponse{
 		ID:         row.ID.String(),
 		Name:       row.Name,
 		ClientType: row.ClientType,
+		Icon:       icon,
 		Enable:     row.Enable,
 		CreatedAt:  row.CreatedAt.Time,
 		UpdatedAt:  row.UpdatedAt.Time,
diff --git a/internal/tts/types.go b/internal/tts/types.go
index b70551a6..19b9ea43 100644
--- a/internal/tts/types.go
+++ b/internal/tts/types.go
@@ -17,6 +17,7 @@ type SpeechProviderResponse struct {
 	ID         string    `json:"id"`
 	Name       string    `json:"name"`
 	ClientType string    `json:"client_type"`
+	Icon       string    `json:"icon,omitempty"`
 	Enable     bool      `json:"enable"`
 	CreatedAt  time.Time `json:"created_at"`
 	UpdatedAt  time.Time `json:"updated_at"`
diff --git a/packages/sdk/src/types.gen.ts b/packages/sdk/src/types.gen.ts
index baf3ae0f..27513ce0 100644
--- a/packages/sdk/src/types.gen.ts
+++ b/packages/sdk/src/types.gen.ts
@@ -1709,6 +1709,7 @@ export type TtsSpeechProviderResponse = {
     client_type?: string;
     created_at?: string;
     enable?: boolean;
+    icon?: string;
     id?: string;
     name?: string;
     updated_at?: string;
diff --git a/spec/docs.go b/spec/docs.go
index 7433d77b..d1806fb0 100644
--- a/spec/docs.go
+++ b/spec/docs.go
@@ -13111,6 +13111,9 @@ const docTemplate = `{
                 "enable": {
                     "type": "boolean"
                 },
+                "icon": {
+                    "type": "string"
+                },
                 "id": {
                     "type": "string"
                 },
diff --git a/spec/swagger.json b/spec/swagger.json
index 2132a109..87ef72bf 100644
--- a/spec/swagger.json
+++ b/spec/swagger.json
@@ -13102,6 +13102,9 @@
                 "enable": {
                     "type": "boolean"
                 },
+                "icon": {
+                    "type": "string"
+                },
                 "id": {
                     "type": "string"
                 },
diff --git a/spec/swagger.yaml b/spec/swagger.yaml
index 8fd0a018..24bc5688 100644
--- a/spec/swagger.yaml
+++ b/spec/swagger.yaml
@@ -2880,6 +2880,8 @@ definitions:
         type: string
       enable:
         type: boolean
+      icon:
+        type: string
       id:
         type: string
       name: