Merge pull request #45 from wasertech/v0.0.21-gm

Version 0.0.21-GM

Author: reuben

.github/workflows/build-and-test.yml

@@ -9,7 +9,7 @@ jobs:
     strategy:
       matrix:
         os: [ubuntu-20.04, macos-10.15, windows-2019]
-        pyver: ['3.6', '3.7', '3.8', '3.9', '3.10']
+        pyver: ['3.7', '3.8', '3.9', '3.10']
       fail-fast: false
     runs-on: ${{ matrix.os }}
     steps:
@@ -48,7 +48,7 @@ jobs:
     runs-on: ubuntu-20.04
     strategy:
       matrix:
-        pyver: ['3.6', '3.7', '3.8', '3.9', '3.10']
+        pyver: ['3.7', '3.8', '3.9', '3.10']
     steps:
       - uses: actions/checkout@v2
       - uses: actions/setup-python@v2

CONTRIBUTING.md

@@ -0,0 +1,44 @@
+# Contribution guidelines
+
+Welcome to the 🐸STT communuty! We are excited to see your interest, and appreciate your support!
+
+This repository is governed by the Contributor Covenant Code of Conduct. For more details, see the [`CODE_OF_CONDUCT.md`](./CODE_OF_CONDUCT.md)
+
+## How to Make a Good Pull Request
+
+Here's some guidelines on how to make a good PR to 🐸STT.
+
+### Bug-fix PR
+
+You've found a bug and you were able to squash it! Great job! Please write a short but clear commit message describing the bug, and how you fixed it. This makes review much easier. Also, please name your branch something related to the bug-fix.
+
+### New Feature PR
+
+You've made some core changes to 🐸STT, and you would like to share them back with the community -- great! First things first: if you're planning to add a feature (not just fix a bug or docs) let the 🐸STT team know ahead of time and get some feedback early. A quick check-in with the team can save time during code-review, and also ensure that your new feature fits into the project.
+
+The 🐸STT codebase is made of many connected parts. There is Python code for training 🐸STT, core C++ code for running inference on trained models, and multiple language bindings to the C++ core so you can use 🐸STT in your favorite language.
+
+Whenever you add a new feature to 🐸STT and what to contribute that feature back to the project, here are some things to keep in mind:
+
+1. You've made changes to the Python code. Make sure you run a linter (described below).
+
+## Requesting review on your PR
+
+Generally, a code owner will be notified of your pull request and will either review it or ask some other code owner for their review. If you'd like to proactively request review as you open the PR, see the the CODE_OWNERS.rst file which describes who's an appropriate reviewer depending on which parts of the code you're changing.
+
+
+## Code linting
+
+We use [pre-commit](https://pre-commit.com/) to manage pre-commit hooks that take care of checking your changes for code style violations. Before committing changes, make sure you have the hook installed in your setup by running, in the virtual environment you use for running the code:
+
+```bash
+$ # Create and activate virtualenv:
+$ python -m venv /tmp/venv
+$ /tmp/venv/bin/activate
+$ # Install package in editable mode:
+$ make dev_install
+$ # Install pre-commit hooks for development:
+$ make install_pre_commit_hooks
+```
+
+This will install a git pre-commit hook which will check your commits and let you know about any style violations that need fixing.

README.md

@@ -1,5 +1,7 @@
 # Coqui STT Model manager
 
+![](./images/cover-1.png)
+
 🐸STT Model Manager is a tool for installing and trying out Coqui STT models from the [Coqui Model Zoo](https://coqui.ai/models/).
 
 # Usage
@@ -21,6 +23,8 @@ $ pip install coqui_stt_model_manager
 $ stt-model-manager
 ```
 
+![](./images/cover-2.png)
+
 # Build and install from source
 
 Build dependencies:
@@ -42,15 +46,7 @@ $ make install
 
 # Development installation (editable, with pre-commit hooks)
 
-```bash
-$ # Create and activate virtualenv:
-$ python -m venv /tmp/venv
-$ /tmp/venv/bin/activate
-$ # Install package in editable mode:
-$ make dev_install
-$ # Install pre-commit hooks for development:
-$ make install_pre_commit_hooks
-```
+See [`CONTRIBUTING`](./CONTRIBUTING.md).
 
 # Running in development mode (watching file changes and rebuilding automatically)
 
@@ -77,4 +73,5 @@ For now you have to reload the browser manually. We'd gladly welcome a PR adding
 ```bash
 $ make package
 $ ll dist/
+$ pip install dist/*.whl
 ```

VERSION

@@ -1 +1 @@
-0.0.20
+0.0.21

coqui_stt_model_manager/modelmanager.py

@@ -1,4 +1,5 @@
 import json
+import logging
 import math
 import os
 import shutil
@@ -13,6 +14,8 @@
 from coqpit import Coqpit
 from xdg import BaseDirectory as xdg
 
+logging.basicConfig(level=logging.DEBUG)
+
 
 @dataclass
 class ModelCard(Coqpit):  # pylint: disable=too-many-instance-attributes
@@ -34,6 +37,13 @@ class ModelCard(Coqpit):  # pylint: disable=too-many-instance-attributes
         default=None, metadata={"help": "path to downloaded scorer, if available"}
     )
 
+    @property
+    def is_installed(self) -> bool:
+        return self.installed
+
+    def __repr__(self):
+        return f"{self.language}/{self.creator}/{self.version}/{self.name}"
+
 
 @dataclass
 class ModelIndex(Coqpit):
@@ -47,31 +57,33 @@ class ModelIndex(Coqpit):
 def _download_one(url: str, dest_path: Path):
     if dest_path.exists():
         yield 100
-        print("Skipped, file already exists")
+        logging.debug("Skipped, file already exists")
         return
 
     response = requests.get(url, stream=True)
     total_length = response.headers.get("Content-Length")
 
     with open(dest_path, "wb") as fout:
         if total_length is None:
-            print(f"Unknown download size for {url}. Downloading...")
+            logging.debug("Unknown download size for %s. Downloading...", url)
             yield 0
             fout.write(response.content)
             yield 100
-            print("Done")
+            logging.debug("Done")
         else:
             total_bytes = int(total_length)
             done_bytes = 0
-            print(f"File is {total_bytes} bytes large for {url}, downloading...")
+            logging.debug(
+                "File is %d bytes large for %s, downloading...", total_bytes, url
+            )
             for chunk in response.iter_content(chunk_size=5 * 2**20):
                 done_bytes += len(chunk)
-                print(f"{done_bytes} out of {total_bytes} downloaded")
+                logging.debug("%d out of %d downloaded", done_bytes, total_bytes)
                 fout.write(chunk)
                 done_pct = math.ceil((done_bytes / total_bytes) * 100)
                 yield done_pct
             yield 100
-            print("Done")
+            logging.debug("Done")
 
 
 class ModelInstallTask(Thread):  # pylint: disable=too-many-instance-attributes
@@ -176,9 +188,17 @@ def __init__(self, install_dir: Optional[Path] = None):
         else:
             self.installed_models = ModelIndex()
             self.persist_index_to_disk()
-        print(f"Installed models: {self.installed_models}")
+        if not self.list_models():
+            logging.debug("No installed models.")
+        else:
+            logging.debug("Installed models:")
+            for model_card in self.list_models():
+                logging.debug(model_card.__repr__())
         self.install_tasks: Dict[str, ModelInstallTask] = {}
 
+    def __repr__(self):
+        return f"ModelManager:{self.model_index_path}"
+
     def read_index_from_disk(self):
         with open(self.model_index_path) as fin:
             self.installed_models = ModelIndex.new_from_dict(json.load(fin))
@@ -188,7 +208,7 @@ def persist_index_to_disk(self):
             fout.write(self.installed_models.to_json())
 
     def list_models(self) -> List[ModelCard]:
-        return self.installed_models.models
+        return [m for m in self.installed_models.models if m.is_installed]
 
     def models_dict(self) -> Dict[str, ModelCard]:
         return {m.name: m for m in self.installed_models.models}
@@ -260,7 +280,7 @@ def download_model(self, model_card: dict):
             scorer_path=output_scorer,
         )
         self.set_install_task_state(install_id, install_task)
-        print("Starting install thread...")
+        logging.debug("Starting install thread...")
         install_task.start()
         return install_id
 
@@ -281,8 +301,9 @@ def maybe_upgrade_protobuf(self):
             m for m in self.list_models() if m.acoustic_path.endswith(".pbmm")
         ]
         if pbmm_models:
-            print(
-                f"Found {len(pbmm_models)} installed protobuf models, upgrading to TFLite if available or removing..."
+            logging.debug(
+                "Found %d installed protobuf models, upgrading to TFLite if available or removing...",
+                len(pbmm_models),
             )
             for model in pbmm_models:
                 self.uninstall_model(model.name)
@@ -296,8 +317,8 @@ def maybe_upgrade_protobuf(self):
                         os.remove(model.acoustic_path)
                     self.download_model(model.to_dict())
                 except:  # pylint: disable=bare-except
-                    print(
-                        f"Couldn't automatically upgrade {model.name}, uninstalling..."
+                    logging.warning(
+                        "Couldn't automatically upgrade %s, uninstalling...", model.name
                     )
                     model_base_path = self.install_dir / model.name
                     shutil.rmtree(model_base_path)

images/cover-1.png

@@ -29,7 +29,7 @@ install_requires =
     webrtcvad==2.0.10
     stt<2.0.0
     requests==2.25.1
-python_requires = >=3.6,<3.11
+python_requires = >=3.7,<3.11
 
 [options.package_data]
 coqui_stt_model_manager = templates/*, static/build/*, static/build/static/js/*, static/build/static/css/*