projects

src/servers/api/server-settings/tts.ts

@@ -54,7 +54,7 @@ ttsRouter.put('/', async (ctx) => {
 });
 
 ttsRouter.post('/voices', async (ctx) => {
-  const body = await ctx.req.json<{ provider: string; url?: string; apiKey?: string }>();
+  const body = await ctx.req.json<{ provider: string; url?: string; apiKey?: string; model?: string }>();
 
   try {
     if (body.provider === 'elevenlabs') {
@@ -67,20 +67,88 @@ ttsRouter.post('/voices', async (ctx) => {
       return ctx.json({ voices: json.voices.map((v) => v.voice_id) });
     }
 
-    // OpenAI-compatible
+    // OpenAI-compatible: try local server first, fallback to HuggingFace
     if (!body.url) return ctx.json({ error: 'URL required' }, 400);
     const headers: Record<string, string> = {};
     if (body.apiKey) headers['Authorization'] = `Bearer ${body.apiKey}`;
-    const res = await fetch(`${body.url.replace(/\/+$/, '')}/v1/audio/voices`, { headers });
-    if (!res.ok) return ctx.json({ error: `Voices fetch failed: ${res.status}` }, 500);
-    const json = (await res.json()) as { voices: string[] };
-    return ctx.json({ voices: json.voices });
+
+    // Try /v1/audio/voices on the local server
+    const localRes = await fetch(`${body.url.replace(/\/+$/, '')}/v1/audio/voices`, { headers }).catch(() => null);
+    if (localRes?.ok) {
+      const json = (await localRes.json()) as { voices: string[] };
+      return ctx.json({ voices: json.voices });
+    }
+
+    // Fallback: get model repo from /v1/models, then list voices from HuggingFace
+    const modelsRes = await fetch(`${body.url.replace(/\/+$/, '')}/v1/models`, { headers }).catch(() => null);
+    if (modelsRes?.ok) {
+      const modelsJson = (await modelsRes.json()) as { data?: { id: string }[] };
+      const repoId = modelsJson.data?.[0]?.id;
+      if (repoId) {
+        const result = await fetchHuggingFaceVoices(repoId);
+        if (result.flat.length > 0) return ctx.json({ voices: result.flat, groups: result.groups });
+      }
+    }
+
+    // Last resort: try model field as HuggingFace repo ID directly
+    if (body.model && body.model.includes('/')) {
+      const result = await fetchHuggingFaceVoices(body.model);
+      if (result.flat.length > 0) return ctx.json({ voices: result.flat, groups: result.groups });
+    }
+
+    return ctx.json({ error: 'Could not fetch voices from server or HuggingFace' }, 500);
   } catch (err) {
     const message = err instanceof Error ? err.message : 'Unknown error';
     return ctx.json({ error: message }, 500);
   }
 });
 
+const VOICE_GROUP_LABELS: Record<string, string> = {
+  af: 'American Female',
+  am: 'American Male',
+  bf: 'British Female',
+  bm: 'British Male',
+  ef: 'Spanish Female',
+  em: 'Spanish Male',
+  ff: 'French Female',
+  hf: 'Hindi Female',
+  hm: 'Hindi Male',
+  if: 'Italian Female',
+  im: 'Italian Male',
+  jf: 'Japanese Female',
+  jm: 'Japanese Male',
+  pf: 'Brazilian Portuguese Female',
+  pm: 'Brazilian Portuguese Male',
+  zf: 'Mandarin Chinese Female',
+  zm: 'Mandarin Chinese Male',
+};
+
+type VoiceGroup = { label: string; voices: string[] };
+
+async function fetchHuggingFaceVoices(repoId: string): Promise<{ flat: string[]; groups: VoiceGroup[] }> {
+  const res = await fetch(`https://huggingface.co/api/models/${repoId}/tree/main/voices`);
+  if (!res.ok) return { flat: [], groups: [] };
+  const files = (await res.json()) as { path: string; type: string }[];
+  const names = new Set<string>();
+  for (const f of files) {
+    if (f.type !== 'file') continue;
+    const name = f.path.replace('voices/', '').replace(/\.(pt|safetensors)$/, '');
+    names.add(name);
+  }
+  const sorted = [...names].sort();
+  const groupMap = new Map<string, string[]>();
+  for (const name of sorted) {
+    const prefix = name.slice(0, 2);
+    if (!groupMap.has(prefix)) groupMap.set(prefix, []);
+    groupMap.get(prefix)!.push(name);
+  }
+  const groups: VoiceGroup[] = [];
+  for (const [prefix, voices] of groupMap) {
+    groups.push({ label: VOICE_GROUP_LABELS[prefix] ?? prefix, voices });
+  }
+  return { flat: sorted, groups };
+}
+
 ttsRouter.post('/test', async (ctx) => {
   const body = await ctx.req.json<TtsConfig>();