v1.1.0
Download Options
| Package | Description | Size |
|---|---|---|
birda-linux-x64.tar.gz |
Linux CPU-only | ~5 MB |
birda-windows-x64.zip |
Windows CPU-only | ~5 MB |
birda-linux-x64-cuda.tar.gz |
Linux with GPU support (bundled CUDA) | ~1550 MB |
birda-windows-x64-cuda.zip |
Windows with GPU support (bundled CUDA) | ~1500 MB |
birda-windows-x64-cuda-setup.exe |
Windows GPU installer (adds to PATH) | ~1000 MB |
GPU Builds
GPU builds include bundled ONNX Runtime, CUDA, and cuDNN libraries - no separate installation required.
Windows Installer: Use the -setup.exe for automatic installation with optional PATH configuration.
Portable: Extract ZIP/tar.gz and run birda directly.
Requirements: NVIDIA GPU with CUDA compute capability 5.0+
See THIRD_PARTY_LICENSES.txt for bundled library licenses.
What's Changed
- Add pre-commit git hooks for code quality by @tphakala in #32
- feat: add comprehensive execution provider selection with explicit flags by @tphakala in #34
- chore: update third-party licenses by @github-actions[bot] in #28
- feat: enhance CLI output with audio decoding feedback and performance metrics by @tphakala in #35
- feat: add RF64/WAV64 support for large audio files by @tphakala in #36
- feat: reimplement CLI output enhancements by @tphakala in #39
- chore: update third-party licenses by @github-actions[bot] in #38
- feat: implement streaming decode pipeline for pipelined inference by @tphakala in #41
- feat(gpu): add inference watchdog to prevent system lockup by @tphakala in #42
- feat: update birdnet-onnx to 2.0.0-rc.4 by @tphakala in #43
- feat: implement BatchInferenceContext for GPU memory efficiency by @tphakala in #44
- feat: add TensorRT warmup to prevent watchdog timeout during engine build by @tphakala in #45
Full Changelog: v1.0.0...v1.1.0
Contributors
tphakala