mobile

src/servers/api/pi/rest.ts

@@ -2,6 +2,7 @@ import type { Context } from 'hono';
 import { createRouter } from '../../create-router';
 import * as storage from './storage';
 import { readApiKeys, readLocalProviders } from '../server-settings/pi-mono';
+import { readSttConfig } from '../server-settings/stt';
 import { listPiModels } from './list-models';
 import { getHomeDir } from '../../data-path';
 import { resolveBaseCwd } from './websocket';
@@ -497,3 +498,42 @@ piRestRouter.post('/pi/sessions/:sessionId/move', async (ctx: Context) => {
     return ctx.json({ error: 'Failed to move session' }, 500);
   }
 });
+
+/**
+ * POST /api/pi/stt
+ * Proxy audio to configured Whisper server for speech-to-text transcription.
+ * Accepts multipart form data with audio file + whisper params.
+ */
+piRestRouter.post('/pi/stt', async (ctx: Context) => {
+  const sttConfig = await readSttConfig();
+  if (!sttConfig?.url) {
+    return ctx.json({ error: 'Whisper not configured — set it up in Settings → Speech to Text' }, 400);
+  }
+
+  const body = await ctx.req.parseBody();
+  const file = body['file'];
+  if (!file || !(file instanceof File)) {
+    return ctx.json({ error: 'file is required' }, 400);
+  }
+
+  const formData = new FormData();
+  formData.append('file', file, 'recording.wav');
+  formData.append('temperature', String(body['temperature'] ?? '0.0'));
+  formData.append('temperature_inc', String(body['temperature_inc'] ?? '0.2'));
+  formData.append('response_format', String(body['response_format'] ?? 'json'));
+
+  try {
+    const res = await fetch(`${sttConfig.url.replace(/\/+$/, '')}/inference`, {
+      method: 'POST',
+      body: formData,
+    });
+    if (!res.ok) {
+      return ctx.json({ error: `Whisper returned ${res.status}` }, 502);
+    }
+    const json = await res.json();
+    return ctx.json(json);
+  } catch (err) {
+    logger.error('STT proxy failed', { error: String(err) });
+    return ctx.json({ error: 'Failed to reach Whisper server' }, 502);
+  }
+});