Use n_quantizers and save codes as uint16 (#12)

* restrict full quantizer usage during inference

* save/load uint16 codes

* add test

* bump version

Author: eeishaan

dac/__init__.py

@@ -1,4 +1,4 @@
-__version__ = "0.0.2"
+__version__ = "0.0.3"
 __model_version__ = "0.0.1"
 import audiotools
 

dac/nn/quantize.py

@@ -171,6 +171,9 @@ def forward(self, z, n_quantizers: int = None):
             n_quantizers = n_quantizers.to(z.device)
 
         for i, quantizer in enumerate(self.quantizers):
+            if self.training is False and i >= n_quantizers:
+                break
+
             z_q_i, commitment_loss_i, codebook_loss_i, indices_i, z_e_i = quantizer(
                 residual
             )

dac/utils/decode.py

@@ -52,7 +52,9 @@ def process(
     """
     if isinstance(generator, torch.nn.DataParallel):
         generator = generator.module
-    audio_signal = AudioSignal(artifacts["codes"], generator.sample_rate)
+    audio_signal = AudioSignal(
+        artifacts["codes"].astype(np.int64), generator.sample_rate
+    )
     metadata = artifacts["metadata"]
 
     # Decode chunks

dac/utils/encode.py

@@ -104,7 +104,7 @@ def process(
     codebook_indices = torch.cat(codebook_indices, dim=0)
 
     return {
-        "codes": codebook_indices.numpy(),
+        "codes": codebook_indices.numpy().astype(np.uint16),
         "metadata": {
             "original_db": input_db,
             "overlap_hop_duration": overlap_hop_duration,

setup.py

@@ -6,7 +6,7 @@
 
 setup(
     name="descript-audio-codec",
-    version="0.0.2",
+    version="0.0.3",
     classifiers=[
         "Intended Audience :: Developers",
         "Natural Language :: English",

tests/test_cli.py

@@ -50,4 +50,24 @@ def test_reconstruction():
         run("decode")
 
 
+def test_compression():
+    # Test encoding
+    input_dir = Path(__file__).parent / "assets" / "input"
+    output_dir = input_dir.parent / "encoded_output_quantizers"
+    args = {"input": str(input_dir), "output": str(output_dir), "n_quantizers": 3}
+    with argbind.scope(args):
+        run("encode")
+
+    # Open .dac file
+    dac_file = output_dir / "sample_0.dac"
+    artifacts = np.load(dac_file, allow_pickle=True)[()]
+    codes = artifacts["codes"]
+
+    # Ensure that the number of quantizers is correct
+    assert codes.shape[1] == 3
+
+    # Ensure that dtype of compression is uint16
+    assert codes.dtype == np.uint16
+
+
 # CUDA_VISIBLE_DEVICES=0 python -m pytest tests/test_cli.py -s