From 3de197ac313e360c70e680625a3a82180f3bcac5 Mon Sep 17 00:00:00 2001
From: imagineer99 <samleejackson0@gmail.com>
Date: Tue, 10 Mar 2026 13:28:49 +0000
Subject: [PATCH] rename: tts model type to audio for broader category support

---
 studio/frontend/src/config/training.ts                    | 8 ++++----
 .../onboarding/components/steps/model-type-step.tsx       | 6 +++---
 studio/frontend/src/features/training/api/models-api.ts   | 2 +-
 .../src/features/training/stores/training-config-store.ts | 2 +-
 studio/frontend/src/hooks/use-hf-dataset-search.ts        | 6 +++---
 studio/frontend/src/types/training.ts                     | 2 +-
 6 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/studio/frontend/src/config/training.ts b/studio/frontend/src/config/training.ts
index 5ae61c5129..e56e5c5391 100644
--- a/studio/frontend/src/config/training.ts
+++ b/studio/frontend/src/config/training.ts
@@ -53,9 +53,9 @@ export const MODEL_TYPES: ReadonlyArray<{
       description: "Image understanding models",
     },
     {
-      value: "tts",
-      label: "TTS",
-      description: "Text-to-speech models",
+      value: "audio",
+      label: "Audio",
+      description: "Audio and speech models",
     },
     {
       value: "embeddings",
@@ -128,6 +128,6 @@ export const DEFAULT_HYPERPARAMS = {
 export const MODEL_TYPE_TO_HF_TASK: Record<ModelType, PipelineType> = {
   text: "text-generation",
   vision: "image-text-to-text",
-  tts: "text-to-speech",
+  audio: "text-to-speech",
   embeddings: "feature-extraction",
 };
diff --git a/studio/frontend/src/features/onboarding/components/steps/model-type-step.tsx b/studio/frontend/src/features/onboarding/components/steps/model-type-step.tsx
index 8fbbe8ec64..1247bab98b 100644
--- a/studio/frontend/src/features/onboarding/components/steps/model-type-step.tsx
+++ b/studio/frontend/src/features/onboarding/components/steps/model-type-step.tsx
@@ -26,19 +26,19 @@ import { useShallow } from "zustand/react/shallow";
 
 const TYPE_ICONS: Record<ModelType, typeof ImageIcon> = {
   vision: ImageIcon,
-  tts: VoiceIcon,
+  audio: VoiceIcon,
   embeddings: Database02Icon,
   text: TextIcon,
 };
 
 const TYPE_TOOLTIPS: Record<ModelType, string> = {
   vision: "Fine-tune models that understand images and text together",
-  tts: "Fine-tune text-to-speech models for voice generation",
+  audio: "Fine-tune text-to-speech and audio models",
   embeddings: "Fine-tune models for semantic search and similarity",
   text: "Fine-tune large language models for text generation",
 };
 
-const COMING_SOON: ModelType[] = ["tts", "embeddings"];
+const COMING_SOON: ModelType[] = ["audio", "embeddings"];
 
 export function ModelTypeStep(): ReactElement {
   const { modelType, setModelType } = useTrainingConfigStore(
diff --git a/studio/frontend/src/features/training/api/models-api.ts b/studio/frontend/src/features/training/api/models-api.ts
index 2ed4645b16..f40f29fda2 100644
--- a/studio/frontend/src/features/training/api/models-api.ts
+++ b/studio/frontend/src/features/training/api/models-api.ts
@@ -64,7 +64,7 @@ export interface ModelConfigResponse {
   is_lora: boolean;
   is_audio?: boolean;
   base_model?: string | null;
-  model_type?: "text" | "vision" | "tts" | "embeddings" | null;
+  model_type?: "text" | "vision" | "audio" | "embeddings" | null;
 }
 
 export interface LocalModelInfo {
diff --git a/studio/frontend/src/features/training/stores/training-config-store.ts b/studio/frontend/src/features/training/stores/training-config-store.ts
index 1ae404cb61..502889ee1e 100644
--- a/studio/frontend/src/features/training/stores/training-config-store.ts
+++ b/studio/frontend/src/features/training/stores/training-config-store.ts
@@ -130,7 +130,7 @@ export const useTrainingConfigStore = create<TrainingConfigStore>()(
             // Use backend-provided model_type when available, otherwise
             // infer from is_vision (temporary until backend ships model_type).
             const inferredModelType: ModelType = modelDetails.model_type
-              ?? (modelDetails.is_vision ? "vision" : modelDetails.is_audio ? "tts" : "text");
+              ?? (modelDetails.is_vision ? "vision" : modelDetails.is_audio ? "audio" : "text");
 
             set({
               ...patch,
diff --git a/studio/frontend/src/hooks/use-hf-dataset-search.ts b/studio/frontend/src/hooks/use-hf-dataset-search.ts
index 3e4ee3b824..9361c7aa32 100644
--- a/studio/frontend/src/hooks/use-hf-dataset-search.ts
+++ b/studio/frontend/src/hooks/use-hf-dataset-search.ts
@@ -118,7 +118,7 @@ const BOOSTED_TASK_CATEGORIES: Record<ModelType, Set<string>> = {
     "image-to-text",
     "image-captioning",
   ]),
-  tts: new Set([
+  audio: new Set([
     "text-to-speech",
     "text-to-audio",
     "automatic-speech-recognition",
@@ -195,7 +195,7 @@ const CURATED_EMPTY_QUERY_DATASET_IDS: Partial<Record<ModelType, string[]>> = {
     "lmms-lab/VQAv2",
     "hezarai/parsynth-ocr-200k",
   ],
-  tts: [
+  audio: [
     "MrDragonFox/Elise",
     "keithito/lj_speech",
     "parler-tts/mls_eng_10k",
@@ -242,7 +242,7 @@ const INCOMPATIBLE_TASKS_BY_MODEL: Record<ModelType, Set<string>> = {
     "audio-to-audio",
     "automatic-speech-recognition",
   ]),
-  tts: new Set([
+  audio: new Set([
     "text-to-image",
     "image-to-image",
     "image-to-video",
diff --git a/studio/frontend/src/types/training.ts b/studio/frontend/src/types/training.ts
index 690929b9a3..205ff656ba 100644
--- a/studio/frontend/src/types/training.ts
+++ b/studio/frontend/src/types/training.ts
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: AGPL-3.0-only - See /studio/LICENSE.AGPL-3.0
 // Copyright © 2025 Unsloth AI
 
-export type ModelType = "vision" | "tts" | "embeddings" | "text";
+export type ModelType = "vision" | "audio" | "embeddings" | "text";
 export type TrainingMethod = "qlora" | "lora" | "full";
 
 export function isAdapterMethod(method: TrainingMethod): boolean {