Spokestack Python Versions Save

What’s Changed

• v0.0.23 (#105) @will-rice
• Pin numpy to 1.19.2 in requirements.txt (#103) @will-rice
• Add PyTorch model abstraction (#100) @will-rice

What’s Changed

• Bump py from 1.9.0 to 1.10.0 (#96) @dependabot
• Bump rsa from 4.6 to 4.7 (#94) @dependabot
• Bump pyyaml from 5.3.1 to 5.4 (#95) @dependabot
• Bump pydantic from 1.7.3 to 1.7.4 (#97) @dependabot
• Bump Dependencies to Latest Version (#93) @will-rice
• Bump pydantic from 1.7.3 to 1.7.4 (#91) @dependabot

What’s Changed

• Increment version to 0.0.21 (#89) @will-rice
• Bump pyyaml from 5.3.1 to 5.4 (#84) @dependabot
• Bump rsa from 4.6 to 4.7 (#83) @dependabot
• Create codeql-analysis.yml (#82) @will-rice
• Bump pygments from 2.6.1 to 2.7.4 in /docs (#80) @dependabot
• Bump jinja2 from 2.11.2 to 2.11.3 in /docs (#81) @dependabot
• Bump py from 1.9.0 to 1.10.0 (#85) @dependabot
• Bump pyyaml from 5.3.1 to 5.4 in /docs (#79) @dependabot
• add tts-lite components (#87) @brentspell
• Update github-deploy.yml (#86) @will-rice
• Add Release Drafter GitHub Action (#78) @will-rice

Features

• Audio input support for sounddevice library.

Features

• Addition of SpokestackKeyword profile to enable users to easily use keyword models.

Features

• Microphone support via PyAudio without the need to install system packages.
• Wheels for Linux, macOS, Windows.

Bug fixes

• Numpy/Cython ValueError caused by version mismatch. Numpy is now pinned to 1.19.2 and will stay in sync with current TensorFlow releases.