sonusai 0.15.4__tar.gz → 0.15.5__tar.gz

{sonusai-0.15.4 → sonusai-0.15.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sonusai
-Version: 0.15.4
+Version: 0.15.5
 Summary: Framework for building deep neural network models for sound, speech, and voice AI
 Home-page: https://aaware.com
 License: GPL-3.0-only
@@ -17,12 +17,12 @@ Classifier: Programming Language :: Python :: 3.11
 Requires-Dist: PyYAML (>=6.0.1,<7.0.0)
 Requires-Dist: aixplain (>=0.2.6,<0.3.0)
 Requires-Dist: dataclasses-json (>=0.6.1,<0.7.0)
-Requires-Dist: deepgram-sdk (>=2.11.0,<3.0.0)
+Requires-Dist: deepgram-sdk (>=3.0.0,<4.0.0)
 Requires-Dist: docopt (>=0.6.2,<0.7.0)
 Requires-Dist: faster-whisper (>=0.10.0,<0.11.0)
 Requires-Dist: ffmpeg (>=1.4,<2.0)
 Requires-Dist: greenlet (>=3.0.1,<4.0.0)
-Requires-Dist: grpcio (==1.59.2)
+Requires-Dist: grpcio (==1.60.0)
 Requires-Dist: jiwer (>=3.0.3,<4.0.0)
 Requires-Dist: keras-tuner (>=1.4.5,<2.0.0)
 Requires-Dist: librosa (>=0.10.1,<0.11.0)
@@ -35,7 +35,7 @@ Requires-Dist: pandas (>=2.1.1,<3.0.0)
 Requires-Dist: pesq (>=0.0.4,<0.0.5)
 Requires-Dist: pyaaware (>=1.5.3,<2.0.0)
 Requires-Dist: pydub (>=0.25.1,<0.26.0)
-Requires-Dist: pystoi (>=0.3.3,<0.4.0)
+Requires-Dist: pystoi (>=0.4.0,<0.5.0)
 Requires-Dist: python-magic (>=0.4.27,<0.5.0)
 Requires-Dist: requests (>=2.31.0,<3.0.0)
 Requires-Dist: resampy (>=0.4.2,<0.5.0)
@@ -45,7 +45,7 @@ Requires-Dist: sh (>=2.0.6,<3.0.0)
 Requires-Dist: soundfile (>=0.12.1,<0.13.0)
 Requires-Dist: sox (>=1.4.1,<2.0.0)
 Requires-Dist: speechrecognition (>=3.10.0,<4.0.0)
-Requires-Dist: tensorflow (>=2.14.0,<3.0.0)
+Requires-Dist: tensorflow (>=2.15.0,<3.0.0)
 Requires-Dist: tensorflow-addons (>=0.23.0,<0.24.0)
 Requires-Dist: tf2onnx (>=1.15.1,<2.0.0)
 Requires-Dist: torch (>=2.1.0,<3.0.0)

{sonusai-0.15.4 → sonusai-0.15.5}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "sonusai"
-version = "0.15.4"
+version = "0.15.5"
 description = "Framework for building deep neural network models for sound, speech, and voice AI"
 authors = ["Chris Eddington <chris@aaware.com>", "Jason Calderwood <jason@aaware.com>"]
 maintainers = ["Chris Eddington <chris@aaware.com>", "Jason Calderwood <jason@aaware.com>"]
@@ -17,20 +17,20 @@ aawscd_probwrite = 'sonusai.aawscd_probwrite:main'
 PyYAML = "^6.0.1"
 aixplain = "^0.2.6"
 dataclasses-json = "^0.6.1"
-deepgram-sdk = "^2.11.0"
+deepgram-sdk = "^3.0.0"
 docopt = "^0.6.2"
 #einops = "^0"
 faster-whisper = "^0.10.0"
 ffmpeg = "^1.4"
 greenlet = "^3.0.1"
-grpcio = "1.59.2"
+grpcio = "1.60.0"
 jiwer = "^3.0.3"
 keras-tuner = "^1.4.5"
 librosa = "^0.10.1"
 lightning = "^2.1.0"
 matplotlib = "^3.8.0"
 onnx = "1.14.1"
-#onnxruntime-gpu = "^1"
+#onnxruntime-gpu = "^1.16.1"
 onnxruntime = "^1.16.1"
 #openai-whisper = "^20231117"
 paho-mqtt = "^1.6.1"
@@ -38,7 +38,7 @@ pandas = "^2.1.1"
 pesq = "^0.0.4"
 pyaaware = "^1.5.3"
 pydub = "^0.25.1"
-pystoi = "^0.3.3"
+pystoi = "^0.4.0"
 python = ">=3.9,<3.12"
 python-magic = "^0.4.27"
 requests = "^2.31.0"
@@ -49,7 +49,7 @@ sh = "^2.0.6"
 soundfile = "^0.12.1"
 sox = "^1.4.1"
 speechrecognition = "^3.10.0"
-tensorflow = "^2.14.0"
+tensorflow = "^2.15.0"
 tensorflow-addons = "^0.23.0"
 tf2onnx = "^1.15.1"
 torch = "^2.1.0"

{sonusai-0.15.4 → sonusai-0.15.5}/sonusai/onnx_predict.py

@@ -223,6 +223,9 @@ def pad_and_predict(feature: Feature,
             model = rt.InferenceSession(model_name, providers=['CPUExecutionProvider'])
 
     predict_arr = np.vstack(predict)
+    # Combine [sequences, batch_size, ...] into [frames, ...]
+    predict_shape = predict_arr.shape
+    predict_arr = np.reshape(predict_arr, [predict_shape[0] * predict_shape[1], *predict_shape[2:]])
     predict_arr, _ = reshape_outputs(predict=predict_arr, timesteps=timesteps)
     predict_arr = predict_arr[:frames, :]