fix(offline): remove inference-path HF_HUB_OFFLINE guards

0.4.3 wrapped every inference body (`generate`, `transcribe`,
`create_voice_clone_prompt`) with `force_offline_if_cached(True, …)` to
prevent lazy HF lookups from hanging when the network drops
mid-inference (#462). That trade broke online users: the guard flips
`huggingface_hub.constants.HF_HUB_OFFLINE` globally, so any legitimate
metadata call the library makes during generation (e.g. revision
resolution via `HfApi().model_info`) now raises:

    Cannot reach https://huggingface.co/api/models/Qwen/Qwen3-TTS-…:
    offline mode is enabled.

Hit by multiple users on 0.4.3 within hours of release. The offline
blast radius is much larger than the original hang it fixed.

This reverts the inference-path guards. Load-path guards stay — those
worked fine in 0.4.2 and aren't the source of the regression. The
`force_offline_if_cached` helper itself is unchanged; tests still pass.

The #462 hang (network dropping mid-inference) remains unaddressed by
this commit and will need a targeted fix that doesn't flip a global
flag — most likely per-call timeouts or library-specific
`local_files_only` arguments, not a process-wide env mutation.

backend/backends/mlx_backend.py

@@ -193,8 +193,6 @@ class MLXTTSBackend:
 
         logger.info("Generating audio for text: %s", text)
 
-        model_name = f"qwen-tts-{self._current_model_size}"
-
         def _generate_sync():
             """Run synchronous generation in thread pool."""
             # MLX generate() returns a generator yielding GenerationResult objects
@@ -220,40 +218,38 @@ class MLXTTSBackend:
                 logger.warning("Regenerating without voice prompt.")
                 ref_audio = None
 
-            # Model is loaded → weights are on disk. Force offline so
+            # Inference runs with the process's default HF_HUB_OFFLINE
-            # lazy tokenizer/config lookups inside mlx_audio don't hang
+            # state. Forcing offline here (previously used to avoid lazy
-            # when the user is disconnected (issue #462).
+            # mlx_audio lookups hanging when the network drops mid-inference,
-            with force_offline_if_cached(True, model_name):
+            # issue #462) regressed online users because libraries make
-                # Check if model supports voice cloning via generate method
+            # legitimate metadata calls during generation.
-                # MLX API may support ref_audio parameter directly
+            try:
-                try:
+                if ref_audio:
-                    # Try with voice cloning parameters if supported
+                    # Check if generate accepts ref_audio parameter
-                    if ref_audio:
+                    import inspect
-                        # Check if generate accepts ref_audio parameter
-                        import inspect
 
-                        sig = inspect.signature(self.model.generate)
+                    sig = inspect.signature(self.model.generate)
-                        if "ref_audio" in sig.parameters:
+                    if "ref_audio" in sig.parameters:
-                            # Generate with voice cloning
+                        # Generate with voice cloning
-                            for result in self.model.generate(text, ref_audio=ref_audio, ref_text=ref_text, lang_code=lang):
+                        for result in self.model.generate(text, ref_audio=ref_audio, ref_text=ref_text, lang_code=lang):
-                                audio_chunks.append(np.array(result.audio))
+                            audio_chunks.append(np.array(result.audio))
-                                sample_rate = result.sample_rate
+                            sample_rate = result.sample_rate
-                        else:
-                            # Fallback: generate without voice cloning
-                            for result in self.model.generate(text, lang_code=lang):
-                                audio_chunks.append(np.array(result.audio))
-                                sample_rate = result.sample_rate
                     else:
-                        # No voice prompt, generate normally
+                        # Fallback: generate without voice cloning
                         for result in self.model.generate(text, lang_code=lang):
                             audio_chunks.append(np.array(result.audio))
                             sample_rate = result.sample_rate
-                except Exception as e:
+                else:
-                    # If voice cloning fails, try without it
+                    # No voice prompt, generate normally
-                    logger.warning("Voice cloning failed, generating without voice prompt: %s", e)
                     for result in self.model.generate(text, lang_code=lang):
                         audio_chunks.append(np.array(result.audio))
                         sample_rate = result.sample_rate
+            except Exception as e:
+                # If voice cloning fails, try without it
+                logger.warning("Voice cloning failed, generating without voice prompt: %s", e)
+                for result in self.model.generate(text, lang_code=lang):
+                    audio_chunks.append(np.array(result.audio))
+                    sample_rate = result.sample_rate
 
             # Concatenate all chunks
             if audio_chunks:
@@ -347,8 +343,6 @@ class MLXSTTBackend:
         """
         await self.load_model_async(model_size)
 
-        progress_model_name = f"whisper-{self.model_size}"
-
         def _transcribe_sync():
             """Run synchronous transcription in thread pool."""
             # MLX Whisper transcription using generate method
@@ -357,11 +351,10 @@ class MLXSTTBackend:
             if language:
                 decode_options["language"] = language
 
-            # Model is loaded → weights are on disk. Force offline so
+            # Inference runs with the process's default HF_HUB_OFFLINE
-            # lazy tokenizer/config lookups don't hang when the user is
+            # state — see the comment in MLXTTSBackend.generate for the
-            # disconnected (issue #462).
+            # regression this revert fixes (issue #462).
-            with force_offline_if_cached(True, progress_model_name):
+            result = self.model.generate(str(audio_path), **decode_options)
-                result = self.model.generate(str(audio_path), **decode_options)
 
             # Extract text from result
             if isinstance(result, str):

backend/backends/pytorch_backend.py

@@ -172,19 +172,17 @@ class PyTorchTTSBackend:
                     # This shouldn't happen in practice, but handle it
                     return {"prompt": cached_prompt}, True
 
-        model_name = f"qwen-tts-{self._current_model_size}"
-
         def _create_prompt_sync():
             """Run synchronous voice prompt creation in thread pool."""
-            # Model is loaded → weights are on disk. Force offline so
+            # Inference runs with the process's default HF_HUB_OFFLINE
-            # lazy tokenizer/config lookups inside qwen_tts don't hang
+            # state. Forcing offline here (issue #462) regressed online
-            # when the user is disconnected (issue #462).
+            # users whose libraries issue legitimate metadata lookups
-            with force_offline_if_cached(True, model_name):
+            # during voice-prompt creation.
-                return self.model.create_voice_clone_prompt(
+            return self.model.create_voice_clone_prompt(
-                    ref_audio=str(audio_path),
+                ref_audio=str(audio_path),
-                    ref_text=reference_text,
+                ref_text=reference_text,
-                    x_vector_only_mode=False,
+                x_vector_only_mode=False,
-                )
+            )
 
         # Run blocking operation in thread pool
         voice_prompt_items = await asyncio.to_thread(_create_prompt_sync)
@@ -227,24 +225,20 @@ class PyTorchTTSBackend:
         # Load model
         await self.load_model_async(None)
 
-        model_name = f"qwen-tts-{self._current_model_size}"
-
         def _generate_sync():
             """Run synchronous generation in thread pool."""
             # Set seed if provided
             if seed is not None:
                 manual_seed(seed, self.device)
 
-            # Model is loaded → weights are on disk. Force offline so
+            # See _create_prompt_sync comment — inference runs with the
-            # lazy tokenizer/config lookups inside qwen_tts don't hang
+            # process's default HF_HUB_OFFLINE state (issue #462).
-            # when the user is disconnected (issue #462).
+            wavs, sample_rate = self.model.generate_voice_clone(
-            with force_offline_if_cached(True, model_name):
+                text=text,
-                wavs, sample_rate = self.model.generate_voice_clone(
+                voice_clone_prompt=voice_prompt,
-                    text=text,
+                language=LANGUAGE_CODE_TO_NAME.get(language, "auto"),
-                    voice_clone_prompt=voice_prompt,
+                instruct=instruct,
-                    language=LANGUAGE_CODE_TO_NAME.get(language, "auto"),
+            )
-                    instruct=instruct,
-                )
             return wavs[0], sample_rate
 
         # Run blocking inference in thread pool to avoid blocking event loop
@@ -342,46 +336,44 @@ class PyTorchSTTBackend:
         """
         await self.load_model_async(model_size)
 
-        progress_model_name = f"whisper-{self.model_size}"
-
         def _transcribe_sync():
             """Run synchronous transcription in thread pool."""
             # Load audio
             audio, _sr = load_audio(audio_path, sample_rate=16000)
 
-            # Model is loaded → weights are on disk. Force offline so
+            # Inference runs with the process's default HF_HUB_OFFLINE
-            # `get_decoder_prompt_ids` and any lazy tokenizer lookups
+            # state — forcing offline here (issue #462) broke online users
-            # don't hang when the user is disconnected (issue #462).
+            # whose `get_decoder_prompt_ids` / tokenizer calls issue
-            with force_offline_if_cached(True, progress_model_name):
+            # legitimate metadata lookups.
-                # Process audio
+            # Process audio
-                inputs = self.processor(
+            inputs = self.processor(
-                    audio,
+                audio,
-                    sampling_rate=16000,
+                sampling_rate=16000,
-                    return_tensors="pt",
+                return_tensors="pt",
+            )
+            inputs = inputs.to(self.device)
+
+            # Generate transcription
+            # If language is provided, force it; otherwise let Whisper auto-detect
+            generate_kwargs = {}
+            if language:
+                forced_decoder_ids = self.processor.get_decoder_prompt_ids(
+                    language=language,
+                    task="transcribe",
                 )
-                inputs = inputs.to(self.device)
+                generate_kwargs["forced_decoder_ids"] = forced_decoder_ids
 
-                # Generate transcription
+            with torch.no_grad():
-                # If language is provided, force it; otherwise let Whisper auto-detect
+                predicted_ids = self.model.generate(
-                generate_kwargs = {}
+                    inputs["input_features"],
-                if language:
+                    **generate_kwargs,
-                    forced_decoder_ids = self.processor.get_decoder_prompt_ids(
+                )
-                        language=language,
-                        task="transcribe",
-                    )
-                    generate_kwargs["forced_decoder_ids"] = forced_decoder_ids
 
-                with torch.no_grad():
+            # Decode
-                    predicted_ids = self.model.generate(
+            transcription = self.processor.batch_decode(
-                        inputs["input_features"],
+                predicted_ids,
-                        **generate_kwargs,
+                skip_special_tokens=True,
-                    )
+            )[0]
-
-                # Decode
-                transcription = self.processor.batch_decode(
-                    predicted_ids,
-                    skip_special_tokens=True,
-                )[0]
 
             return transcription.strip()
 

backend/backends/qwen_custom_voice_backend.py

@@ -186,7 +186,6 @@ class QwenCustomVoiceBackend:
         await self.load_model_async(None)
 
         speaker = voice_prompt.get("preset_voice_id") or QWEN_CV_DEFAULT_SPEAKER
-        model_name = f"qwen-custom-voice-{self._current_model_size}"
 
         def _generate_sync():
             if seed is not None:
@@ -206,11 +205,11 @@ class QwenCustomVoiceBackend:
             if instruct:
                 kwargs["instruct"] = instruct
 
-            # Model is loaded → weights are on disk. Force offline so
+            # Inference runs with the process's default HF_HUB_OFFLINE
-            # lazy tokenizer/config lookups inside qwen_tts don't hang
+            # state. Forcing offline here (issue #462) regressed online
-            # when the user is disconnected (issue #462).
+            # users whose libraries issue legitimate metadata lookups
-            with force_offline_if_cached(True, model_name):
+            # during generation.
-                wavs, sample_rate = self.model.generate_custom_voice(**kwargs)
+            wavs, sample_rate = self.model.generate_custom_voice(**kwargs)
             return wavs[0], sample_rate
 
         audio, sample_rate = await asyncio.to_thread(_generate_sync)