CorentinJ/Real-Time-Voice-Cloning · critical · Exception

No synthesizer models found in %s

Error message

No synthesizer models found in %s

What it means

Raised by Ui.populate_models (toolbox/ui.py) when models_dir.glob("*/synthesizer.pt") finds no match. Unlike the encoder (hard requirement, error 8) and the vocoder (optional, falls back to Griffin-Lim), the synthesizer has no built-in fallback, so the toolbox aborts when it cannot find <models_dir>/<name>/synthesizer.pt.

Source

Thrown at toolbox/ui.py:344

    @property
    def current_synthesizer_fpath(self):
        return self.synthesizer_box.itemData(self.synthesizer_box.currentIndex())

    @property
    def current_vocoder_fpath(self):
        return self.vocoder_box.itemData(self.vocoder_box.currentIndex())

    def populate_models(self, models_dir: Path):
        # Encoder
        encoder_fpaths = list(models_dir.glob("*/encoder.pt"))
        if len(encoder_fpaths) == 0:
            raise Exception("No encoder models found in %s" % models_dir)
        self.repopulate_box(self.encoder_box, [(f.parent.name, f) for f in encoder_fpaths])

        # Synthesizer
        synthesizer_fpaths = list(models_dir.glob("*/synthesizer.pt"))
        if len(synthesizer_fpaths) == 0:
            raise Exception("No synthesizer models found in %s" % models_dir)
        self.repopulate_box(self.synthesizer_box, [(f.parent.name, f) for f in synthesizer_fpaths])

        # Vocoder
        vocoder_fpaths = list(models_dir.glob("*/vocoder.pt"))
        vocoder_items = [(f.parent.name, f) for f in vocoder_fpaths] + [("Griffin-Lim", None)]
        self.repopulate_box(self.vocoder_box, vocoder_items)

    @property
    def selected_utterance(self):
        return self.utterance_history.itemData(self.utterance_history.currentIndex())

    def register_utterance(self, utterance: Utterance):
        self.utterance_history.blockSignals(True)
        self.utterance_history.insertItem(0, utterance.name, utterance)
        self.utterance_history.setCurrentIndex(0)
        self.utterance_history.blockSignals(False)

        if len(self.utterance_history) > self.max_saved_utterances:

View on GitHub (pinned to 890f3a0318)

Solutions

1. Download the pretrained synthesizer and place it at <models_dir>/<model_name>/synthesizer.pt.
2. Confirm the glob matches: ls <models_dir>/*/synthesizer.pt — at least one result is required.
3. If you only meant to use the encoder, note the toolbox cannot start without a synthesizer; use demo_cli.py-style flows or place the pretrained synthesizer.

Example fix

# before
python demo_toolbox.py -m ~/models  # ~/models has encoder.pt but no */synthesizer.pt -> raises

# after
mkdir -p ~/models/pretrained && mv ~/Downloads/synthesizer.pt ~/models/pretrained/
python demo_toolbox.py -m ~/models

from pathlib import Path

def synthesizer_model_present(models_dir: Path) -> bool:
    return any(models_dir.glob("*/synthesizer.pt"))

try:
    ui.populate_models(models_dir)
except Exception as e:
    if "synthesizer models" in str(e):
        raise SystemExit(f"Missing synthesizer: place <models>/<name>/synthesizer.pt") from e
    raise

When it happens

Trigger: Launching demo_toolbox.py where the models root has an encoder.pt but no synthesizer.pt in any subdirectory: synthesizer never downloaded, or the -m argument points at a dir containing only encoder/vocoder checkpoints.

Common situations: Downloading only the pretrained encoder initially; copying synthesizer.pt to the wrong nesting level (must be one subdirectory deep); mixing models from different roots after a directory reorganization.

Related errors

• No encoder models found in %s
• No speakers found. Make sure you are pointing to the directo
• Model was not loaded. Call load_model() before inference.
• No visdom server detected. Run the command \"visdom\" in you
• `hparams.synthesis_batch_size` must be evenly divisible by n

AI-assisted analysis of CorentinJ/Real-Time-Voice-Cloning@890f3a0318 (2026-08-15). Data as JSON: /api/errors/d160d81e923c9481. Report an issue: GitHub.