Version 0.9.2

.github/workflows/pytest.yml

@@ -10,20 +10,23 @@ jobs:
   test:
     runs-on: ubuntu-latest
 
+    strategy:
+      matrix:
+        python-version: ["3.10", "3.11", "3.12"]
+
     steps:
     - name: Checkout code
       uses: actions/checkout@v3
 
-    - name: Set up Python
+    - name: Set up Python ${{ matrix.python-version }}
       uses: actions/setup-python@v3
       with:
-        python-version: '3.10'
+        python-version: ${{ matrix.python-version }}
 
     - name: Install apt dependencies
       run: |
-        sudo add-apt-repository ppa:savoury1/ffmpeg4
         sudo apt-get update
-        sudo apt-get -y install ffmpeg libportaudio2=19.6.0-1.1
+        sudo apt-get -y install ffmpeg libportaudio2
 
     - name: Install pip dependencies
       run: |

.github/workflows/quick-runs.yml

@@ -14,7 +14,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.8", "3.10"]
+        python-version: ["3.10", "3.11", "3.12"]
 
     steps:
     - uses: actions/checkout@v3
@@ -31,9 +31,8 @@ jobs:
         wget --no-verbose --show-progress --continue -O rttms/ES2002b_long.rttm https://raw.githubusercontent.com/pyannote/AMI-diarization-setup/main/only_words/rttms/train/ES2002b.rttm
     - name: Install apt dependencies
       run: |
-        sudo add-apt-repository ppa:savoury1/ffmpeg4
         sudo apt-get update
-        sudo apt-get -y install ffmpeg libportaudio2=19.6.0-1.1 sox
+        sudo apt-get -y install ffmpeg libportaudio2 sox
     - name: Install pip dependencies
       run: |
         python -m pip install --upgrade pip

README.md

@@ -284,21 +284,27 @@ Obtain overlap-aware speaker embeddings from a microphone stream:
 ```python
 import rx.operators as ops
 import diart.operators as dops
-from diart.sources import MicrophoneAudioSource
+from diart.sources import MicrophoneAudioSource, FileAudioSource
 from diart.blocks import SpeakerSegmentation, OverlapAwareSpeakerEmbedding
 
 segmentation = SpeakerSegmentation.from_pretrained("pyannote/segmentation")
 embedding = OverlapAwareSpeakerEmbedding.from_pretrained("pyannote/embedding")
-mic = MicrophoneAudioSource()
+
+source = MicrophoneAudioSource()
+# To take input from file:
+# source = FileAudioSource("<filename>", sample_rate=16000)
+
+# Make sure the models have been trained with this sample rate
+print(source.sample_rate)
 
 stream = mic.stream.pipe(
     # Reformat stream to 5s duration and 500ms shift
-    dops.rearrange_audio_stream(sample_rate=segmentation.model.sample_rate),
+    dops.rearrange_audio_stream(sample_rate=source.sample_rate),
     ops.map(lambda wav: (wav, segmentation(wav))),
     ops.starmap(embedding)
 ).subscribe(on_next=lambda emb: print(emb.shape))
 
-mic.read()
+source.read()
 ```
 
 Output:

requirements.txt

@@ -1,5 +1,5 @@
-numpy>=1.20.2
-matplotlib>=3.3.3,<3.6.0
+numpy>=1.20.2,<2.0.0
+matplotlib>=3.3.3,<4.0.0
 rx>=3.2.0
 scipy>=1.6.0
 sounddevice>=0.4.2

setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name=diart
-version=0.9.1
+version=0.9.2
 author=Juan Manuel Coria
 description=A python framework to build AI for real-time speech
 long_description=file: README.md
@@ -20,8 +20,8 @@ package_dir=
     =src
 packages=find:
 install_requires=
-    numpy>=1.20.2
-    matplotlib>=3.3.3,<3.6.0
+    numpy>=1.20.2,<2.0.0
+    matplotlib>=3.3.3,<4.0.0
     rx>=3.2.0
     scipy>=1.6.0
     sounddevice>=0.4.2