fix(voxtral): Audio als temp-WAV-Pfad an Processor (statt rohem Array)

VoxtralProcessor.apply_transcription_request verlangt bei rohen Arrays ein 'format'. Fix: Buffer in ein temp-WAV (PCM_16, 16kHz) schreiben und den Pfad uebergeben — Processor liest Format+Samplerate selbst. Temp-Datei wird nach dem Transkribieren geloescht.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

xtts/voxtral/bridge.py

@@ -29,11 +29,13 @@ import base64
 import json
 import logging
 import os
+import tempfile
 import time
 from dataclasses import dataclass, field
 from typing import Optional
 
 import numpy as np
+import soundfile as sf
 import websockets
 
 logging.basicConfig(
@@ -102,13 +104,20 @@ class VoxtralRunner:
         logger.info("Voxtral geladen in %.1fs", time.time() - t0)
 
     def _transcribe_blocking(self, audio_f32: np.ndarray, language: str) -> str:
-        # ⚠️ VERIFY: exakte Voxtral-Transformers-API gegen die HF-Modelcard.
         import torch
         proc, model = self.processor, self.model
         if proc is None or model is None or audio_f32.size == 0:
             return ""
+        # VoxtralProcessor verlangt bei rohen Arrays ein 'format'. Robuster:
+        # in ein temp-WAV schreiben und den PFAD uebergeben — der Processor liest
+        # Format + Samplerate selbst, kein 'format'-Argument noetig.
+        wav_path = None
+        try:
+            with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as tf:
+                wav_path = tf.name
+            sf.write(wav_path, audio_f32, 16000, subtype="PCM_16")
             inputs = proc.apply_transcription_request(
-            language=language, audio=audio_f32, model_id=VOXTRAL_MODEL, sampling_rate=16000,
+                language=language, audio=wav_path, model_id=VOXTRAL_MODEL,
             )
             inputs = inputs.to(VOXTRAL_DEVICE, dtype=torch.bfloat16)
             with torch.no_grad():
@@ -116,6 +125,12 @@ class VoxtralRunner:
             trimmed = outputs[:, inputs.input_ids.shape[1]:]
             text = proc.batch_decode(trimmed, skip_special_tokens=True)
             return (text[0] if text else "").strip()
+        finally:
+            if wav_path:
+                try:
+                    os.unlink(wav_path)
+                except Exception:
+                    pass
 
     async def transcribe(self, audio_f32: np.ndarray, language: str) -> str:
         loop = asyncio.get_running_loop()