Merge pull request #77 from sensein/release_040

Release 040

.github/ISSUE_TEMPLATE/general.yaml

@@ -14,4 +14,4 @@ body:
     description: Please, provide a clear and concise piece of information.
     placeholder: Please, provide a clear and concise piece of information.
   validations:
-    required: false
+    required: false

.github/PULL_REQUEST_TEMPLATE/advanced.md

@@ -23,4 +23,4 @@ Include any relevant screenshots.
 ## Checklist:
 - [ ] My code follows the code style of this project.
 - [ ] I have added tests to cover my changes.
-- [ ] All new and existing tests passed.
+- [ ] All new and existing tests passed.

.github/PULL_REQUEST_TEMPLATE/basic.md

@@ -1,4 +1,4 @@
 # Basic Pull Request Template
 
 ## Description
-Briefly describe the changes you've made.
+Briefly describe the changes you've made.

.github/dependabot.yaml

@@ -5,23 +5,23 @@
 version: 2
 updates:
   # Configuration for Python package updates using pip
-  - package-ecosystem: pip
-    directory: "/"
-    schedule:
-      interval: monthly
-    open-pull-requests-limit: 5
-    groups:
-      production-dependencies:
-        dependency-type: production
-      development-dependencies:
-        dependency-type: development
-        
+- package-ecosystem: pip
+  directory: /
+  schedule:
+    interval: monthly
+  open-pull-requests-limit: 5
+  groups:
+    production-dependencies:
+      dependency-type: production
+    development-dependencies:
+      dependency-type: development
+
   # Configuration for GitHub Actions updates
-  - package-ecosystem: github-actions
-    directory: "/"
-    schedule:
-      interval: monthly
-    open-pull-requests-limit: 5
+- package-ecosystem: github-actions
+  directory: /
+  schedule:
+    interval: monthly
+  open-pull-requests-limit: 5
 
 # Enable automatic security updates
-security_updates: true
+security_updates: true

.github/workflows/docs.yaml

@@ -16,11 +16,11 @@ jobs:
       fail-fast: false
       matrix:
         os: [ubuntu-latest]
-        python-version: ["3.10"]
+        python-version: ['3.10']
         include:
-          - os: ubuntu-latest
-            python-version: "3.10"
-            deploy: true # Only deploy from this configuration
+        - os: ubuntu-latest
+          python-version: '3.10'
+          deploy: true   # Only deploy from this configuration
     outputs:
       deploy: ${{ steps.set-deploy-output.outputs.deploy }}
     steps:
@@ -59,11 +59,11 @@ jobs:
       fail-fast: false
       matrix:
         os: [ubuntu-latest] # For demonstration, other OSes are commented out
-        python-version: ["3.10"]
+        python-version: ['3.10']
         include:
-          - os: ubuntu-latest
-            python-version: "3.10"
-            deploy: true # Only deploy from this configuration
+        - os: ubuntu-latest
+          python-version: '3.10'
+          deploy: true   # Only deploy from this configuration
     steps:
     - uses: actions/checkout@v4
     - uses: actions/download-artifact@v3

.github/workflows/publish.yaml

@@ -10,7 +10,7 @@ jobs:
     strategy:
       matrix:
         os: [ubuntu-latest]
-        python-version: ["3.10"]
+        python-version: ['3.10']
     steps:
     - uses: actions/checkout@v4
       with:
@@ -41,7 +41,3 @@ jobs:
         POETRY_REQUESTS_TIMEOUT: 120
       run: poetry publish
       shell: bash
-
-
-
-

.github/workflows/release.yaml

@@ -3,7 +3,7 @@ name: Auto-release
 on:
   push:
     branches:
-      - main
+    - main
 
 env:
   AUTO_VERSION: v11.1.2
@@ -13,31 +13,30 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-      - uses: actions/checkout@v4
+    - uses: actions/checkout@v4
 
-      - name: Prepare repository
+    - name: Prepare repository
         # Fetch full git history and tags
-        run: git fetch --unshallow --tags
+      run: git fetch --unshallow --tags
 
-      - name: Unset header
+    - name: Unset header
         # checkout@v2 adds a header that makes branch protection report errors
         # because the Github action bot is not a collaborator on the repo
-        run: git config --local --unset http.https://github.com/.extraheader
-
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.10"
-
-      - name: Download auto
-        run: |
-          auto_download_url="$(curl -fsSL https://api.github.com/repos/intuit/auto/releases/tags/$AUTO_VERSION | jq -r '.assets[] | select(.name == "auto-linux.gz") | .browser_download_url')"
-          wget -O- "$auto_download_url" | gunzip > ~/auto
-          chmod a+x ~/auto
-
-      - name: Create release
-        run: |
-          ~/auto shipit -vv
-        env:
-          GH_TOKEN: ${{ secrets.AUTO_ORG_TOKEN }}
-
+      run: git config --local --unset http.https://github.com/.extraheader
+
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: '3.10'
+
+    - name: Download auto
+      run: |
+        auto_download_url="$(curl -fsSL https://api.github.com/repos/intuit/auto/releases/tags/$AUTO_VERSION | jq -r '.assets[] | select(.name == "auto-linux.gz") | .browser_download_url')"
+        wget -O- "$auto_download_url" | gunzip > ~/auto
+        chmod a+x ~/auto
+
+    - name: Create release
+      run: |
+        ~/auto shipit -vv
+      env:
+        GH_TOKEN: ${{ secrets.AUTO_ORG_TOKEN }}

.gitignore

@@ -167,3 +167,7 @@ cython_debug/
 .idea/
 
 data/
+
+# Speechbrain models
+pretrained_models
+src/tests/audio/tasks/pretrained_models/

.pre-commit-config.yaml

@@ -19,6 +19,7 @@ repos:
     args: [--ignore-missing-imports]
     additional_dependencies:
     - pydantic>=1.10.4
+    - types-requests
 - repo: https://github.com/macisamuele/language-formatters-pre-commit-hooks
   rev: v2.12.0
   hooks:

ACTION_PLAN.md

@@ -1,9 +1,9 @@
 # Action Plan for Senselab
 
-Senselab is a Python package designed to streamline the processing and analysis of behavioral data, including voice and speech, text, and video. Our goal is to implement robust and reproducible methodologies. 
+Senselab is a Python package designed to streamline the processing and analysis of behavioral data, including voice and speech, text, and video. Our goal is to implement robust and reproducible methodologies.
 
 This action plan outlines our initial functionalities and integrations for Senselab. As we progress, we will continue to enhance and expand these capabilities to support more advanced and comprehensive behavioral data analysis.
-Anyone should feel free to suggest more features and methods. 
+Anyone should feel free to suggest more features and methods.
 
 For an updated project progress, please see the [Project Board](https://github.com/orgs/sensein/projects/45).
 

CONTRIBUTING.md

@@ -7,7 +7,7 @@ Note that a code of conduct applies to all spaces managed by the `senselab` proj
 Please use the following workflow when contributing:
 
 0. **Install poetry and poetry-dynamic-versioning plugin for dynamic versioning**: 
-  - ```pipx install poetry```
+  - ```pipx install poetry==1.7.1```
   - ```pipx inject poetry "poetry-dynamic-versioning[plugin]"```
 1. **Create an issue**: Use GitHub to create an issuel, assign it to yourself (and any collaborators) and, if you have access, add it to the [Project board](https://github.com/orgs/sensein/projects/45).
 2. **Create a branch**: Use GitHub's "Create a branch" button from the issue page to generate a branch associated with the issue.
@@ -17,7 +17,8 @@ Please use the following workflow when contributing:
     - ```git fetch origin```
     - ```git checkout <branch-name>```
 5. **Install all required dependencies**:
-  ```poetry install --with dev,docs```
+  - ```poetry run pip install iso-639```
+  - ```poetry install --with dev,docs```
 6. **Install pre-commit hooks**:
   ```poetry run pre-commit install```
 7. **Work locally on the issue branch.**

FEATURES.tmp.md

@@ -1,12 +1,12 @@
-# Functionalities 
-This file is here just as a support for development. 
+# Functionalities
+This file is here just as a support for development.
 
 AUDIO
 
-[TODO]: 
+[TODO]:
 - speech to text
     1. to transcribe speech into text
-        - INPUT: 
+        - INPUT:
             1. a datasets object with the audio recordings in the "audio" column
             2. the audio column (default = "audio")
             3. the speech to text service to use (including the name, the version, the revision, and - for some services only and sometimes it's optional - the language of the transcription model we want to use)
@@ -15,7 +15,7 @@ AUDIO
             2. organize the dataset into batches
         - PROCESSING:
             1. transcribe the dataset
-        - POSTPROCESSING: 
+        - POSTPROCESSING:
             1. formatting the transcripts to follow a standard organization
         - OUTPUT:
             1. a new dataset including only the transcripts of the audios in a standardized json format (plus an index?)
@@ -25,14 +25,13 @@ AUDIO
             3. test the language is supported (and the tool handles errors)
 
     2. to compute word error rate
-        - INPUT: 
-            1. a dataset object with the "transcript" and the "groundtruth" columns 
+        - INPUT:
+            1. a dataset object with the "transcript" and the "groundtruth" columns
             2. a service with a name (default is jitter)
         - PROCESSING:
             1. computing the per-row WER between the 2 columns
-        - OUTPUT: 
+        - OUTPUT:
             1. a dataset with the "WER" column
         - TESTS:
             1. test input errors (a field is missing, fields missing, the 2 columns don't contain strings)
             2. test output is ok
-

pyproject.toml

@@ -45,6 +45,7 @@ torch-audiomentations = "^0.11.1"
 sentence-transformers = "^2.7.0"
 jiwer = "^3.0.4"
 speechbrain = "^1.0.0"
+pyav = "^12.1.0"
 pyannote-audio = "^3.2.0"
 pycountry = "^24.6.1"
 types-requests = "^2.32"
@@ -149,7 +150,7 @@ skip = [
   "poetry.lock",
   "docs_style/pdoc-theme/syntax-highlighting.css"
 ]
-ignore-words-list = ["senselab", "nd", "astroid", "wil"]
+ignore-words-list = ["senselab", "nd", "astroid", "wil", "SER"]
 
 [build-system]
 requires = ["poetry-core>=1.0.0", "poetry-dynamic-versioning>=1.0.0,<2.0.0"]

src/senselab/audio/data_structures/audio.py

@@ -6,6 +6,7 @@
 ease of maintaining the codebase and offering consistent public APIs.
 """
 
+import os
 import uuid
 from typing import Dict, List, Optional, Tuple, Union
 
@@ -35,7 +36,7 @@ class Audio(BaseModel):
 
     waveform: torch.Tensor
     sampling_rate: int
-    orig_path_or_id: Optional[str] = Field(default_factory=lambda: str(uuid.uuid4()))
+    orig_path_or_id: Optional[str] = None
     metadata: Dict = Field(default={})
     model_config = {"arbitrary_types_allowed": True}
 
@@ -57,7 +58,7 @@ def convert_to_tensor(
         if len(temporary_tensor.shape) == 1:
             # make the audio data [channels=1, samples]
             temporary_tensor = temporary_tensor.unsqueeze(0)
-        return temporary_tensor
+        return temporary_tensor.to(torch.float32)
 
     @classmethod
     def from_filepath(cls, filepath: str, metadata: Dict = {}) -> "Audio":
@@ -71,6 +72,21 @@ def from_filepath(cls, filepath: str, metadata: Dict = {}) -> "Audio":
 
         return cls(waveform=array, sampling_rate=sampling_rate, orig_path_or_id=filepath, metadata=metadata)
 
+    def generate_path(self) -> str:
+        """Generate a path like string for this Audio.
+
+        Generates a path like string for the Audio by either utilizing the orig_path_or_id, checking
+        if it is a path (has an extension), otherwise using the id if orig_path_or_id is not an ID
+        and giving an extension and relative to the current directory.
+        """
+        if self.orig_path_or_id:
+            if os.path.splitext(self.orig_path_or_id)[-1].lower():
+                return self.orig_path_or_id
+            else:
+                return f"{self.orig_path_or_id}"
+        else:
+            return f"{self.id()}"
+
     def id(self) -> str:
         """Generate a unique identifier for the Audio.
 

src/senselab/audio/tasks/classification/__init__.py

@@ -0,0 +1 @@
+"""This module defines the APIs for doing generalized audio classification tasks."""