|
Download README.md from tphakala/BirdNET-Plus-V3.0: direct link, hf CLI and curl.
- Browser
- Download file 9.45 kB
-
https://huggingface.co/tphakala/BirdNET-Plus-V3.0/resolve/main/README.md
- Command line
-
hf download hf://tphakala/BirdNET-Plus-V3.0/README.md
-
curl -L -o README.md https://huggingface.co/tphakala/BirdNET-Plus-V3.0/resolve/main/README.md
9.45 kB
| pretty_name: BirdNET+ V3.0 (GPU-native full + regional catalog) | |
| license: apache-2.0 | |
| tags: | |
| - bioacoustics | |
| - birdnet | |
| - bird | |
| - vocalization | |
| - audio-classification | |
| - onnx | |
| library_name: onnx | |
| # BirdNET+ V3.0 (GPU-native full + regional catalog) | |
| BirdNET+ V3.0 acoustic classifier, re-exported to a | |
| **GPU-native ONNX** graph, plus a catalog of **region-specific slices** that are | |
| much smaller while staying numerically identical to the full model on the | |
| species they keep. | |
| ## Licence and terms | |
| Each release came with its own upstream terms, listed here with its licence. The | |
| repo-level licence tag is the current release's. | |
| - **BirdNET+ V3.0** (`birdnet-plus-v3.0`), Apache-2.0: the Apache-2.0 model licence and acceptable-use guidance of the BirdNET+ V3.0 release (Zenodo, https://doi.org/10.5281/zenodo.24162518), `LICENSE-v3.0.md` and `ACCEPTABLE_USE-v3.0.md` at the repo root, unchanged from upstream. | |
| ## Origin and attribution | |
| **Powered by BirdNET** (https://birdnet.cornell.edu/). BirdNET+ V3.0 is developed | |
| by the **K. Lisa Yang Center for Conservation Bioacoustics (Cornell University)**, | |
| **Chemnitz University of Technology**, and the **Museum fuer Naturkunde Berlin**. | |
| It builds on Kahl et al. (2021) and Lasseck (2018). Architecture of this release: | |
| EfficientNetV2-S backbone, three-head classifier, 11,725 classes. | |
| The GPU-native conversion and the regional slicing in this repo are derived work | |
| under Apache-2.0. | |
| ## Releases in this repo | |
| The current release is **BirdNET+ V3.0** (`birdnet-plus-v3.0`). | |
| ## What "GPU-native" means | |
| The BirdNET team's ONNX export computes its log-mel front-end with an in-graph | |
| ONNX `STFT` op. ONNX Runtime has no CUDA/TensorRT kernel for `STFT`, and OpenVINO | |
| cannot translate it at all, so the stock export cannot run on GPU or OpenVINO. In | |
| these models the `STFT` is replaced with a mathematically equivalent `Conv1d` | |
| (framing + windowed DFT), so the graph holds no op those runtimes lack. | |
| Every file was run on **ONNX Runtime CPU** (x86-64 and a Raspberry Pi 5) and on **OpenVINO** at f32 (CPU on both, and an Intel integrated GPU), and its scores were compared with the x86-64 ONNX Runtime scores of the same file; the manifest's `supported` and `verified_on` follow those runs. OpenVINO must run at f32: its default f16 hint gives wrong scores on the Cortex-A76 and on Intel GPUs. It is not recommended: its margin over ONNX Runtime CPU was 7% at most where it was measured. The CUDA and TensorRT entries were not run for this release. One model serves CPU and GPU; there is no separate CPU/GPU | |
| build. | |
| Input: raw audio `[batch, 160000]` (**5 s @ 32 kHz**, matching Perch v2 so a host | |
| can feed both classifiers the same buffer). Outputs: `predictions [batch, 11,725]` | |
| (a static dimension, sigmoid already applied) and `embeddings [batch, 1280]`. | |
| ## Variants and hardware | |
| | precision | best for | | |
| |---|---| | |
| | `-fp32-` | **run this**: ONNX Runtime and OpenVINO (BirdNET-Go runs v3.0 at f32 on every OpenVINO device) | | |
| This release ships FP32 only. For low-RAM CPU devices the right choice is a | |
| **regional FP32** model. | |
| int8-arm is not shipped for v3.0: the backbone is Conv-dominated and int8-arm only | |
| quantizes MatMul/Gemm weights, so it barely helps here. | |
| ## Full model | |
| | file | classes | MB | | |
| |---|---:|---:| | |
| | `full/birdnet-plus-v3.0-fp32-b1.onnx` | 11,725 | 361 | | |
| | `full/birdnet-plus-v3.0-labels-b1.txt` | 11,725 | - | | |
| ## Why regional models | |
| The three-head classifier holds a large share of the model's parameters, so | |
| restricting it to a region's species set cuts memory and download size. Each tile | |
| is **numerically identical** to the full model on the species it keeps (max abs | |
| difference vs the full model < 1e-6); the only change is that out-of-region | |
| species are not emitted, which at a fixed monitoring location is exactly what you | |
| want. Region species sets come from the **BirdNET Geomodel v3.0.2** range filter | |
| (species by occurrence probability, up to a per-tile cap, plus a cosmopolitan core), | |
| using the same tiles as the Perch v2 regional catalog. The cap is 800 for most | |
| temperate tiles and higher for species-rich ones. Every tile also contains Homo sapiens and the release's non-animal sound classes, whatever its range filter says. | |
| A species the geomodel names differently from the classifier (a moved genus, a | |
| changed spelling) is matched through a reviewed rename table, so the tiles keep | |
| species an exact name match would have dropped. | |
| ## Regional catalog | |
| Each tile ships an fp32 model + labels + indices + `coverage.png` + | |
| metadata. `regional/groups.json` lists the continent grouping. | |
| ### Europe | |
| | region | coverage | classes | fp32 MB | | |
| |---|---|---:|---:| | |
| | `azores` | <img src="regional/azores/coverage.png" width="210"> | 242 | 113.7 | | |
| | `baltics` | <img src="regional/baltics/coverage.png" width="210"> | 460 | 118.4 | | |
| | `british-isles` | <img src="regional/british-isles/coverage.png" width="210"> | 579 | 121.0 | | |
| | `canary-islands` | <img src="regional/canary-islands/coverage.png" width="210"> | 412 | 117.4 | | |
| | `central-europe` | <img src="regional/central-europe/coverage.png" width="210"> | 667 | 122.9 | | |
| | `eastern-europe` | <img src="regional/eastern-europe/coverage.png" width="210"> | 540 | 120.1 | | |
| | `iberia` | <img src="regional/iberia/coverage.png" width="210"> | 649 | 122.5 | | |
| | `iceland` | <img src="regional/iceland/coverage.png" width="210"> | 409 | 117.3 | | |
| | `madeira` | <img src="regional/madeira/coverage.png" width="210"> | 274 | 114.4 | | |
| | `nordic` | <img src="regional/nordic/coverage.png" width="210"> | 443 | 118.0 | | |
| | `southern-europe` | <img src="regional/southern-europe/coverage.png" width="210"> | 638 | 122.2 | | |
| | `svalbard` | <img src="regional/svalbard/coverage.png" width="210"> | 298 | 114.9 | | |
| | `western-palearctic` | <img src="regional/western-palearctic/coverage.png" width="210"> | 1418 | 139.0 | | |
| ### Asia | |
| | region | coverage | classes | fp32 MB | | |
| |---|---|---:|---:| | |
| | `china-north-central` | <img src="regional/china-north-central/coverage.png" width="210"> | 791 | 125.5 | | |
| | `china-northeast` | <img src="regional/china-northeast/coverage.png" width="210"> | 688 | 123.3 | | |
| | `china-southeast` | <img src="regional/china-southeast/coverage.png" width="210"> | 1178 | 133.9 | | |
| | `china-southwest` | <img src="regional/china-southwest/coverage.png" width="210"> | 1228 | 134.9 | | |
| | `himalaya` | <img src="regional/himalaya/coverage.png" width="210"> | 1226 | 134.9 | | |
| | `indo-gangetic` | <img src="regional/indo-gangetic/coverage.png" width="210"> | 1199 | 134.3 | | |
| | `japan` | <img src="regional/japan/coverage.png" width="210"> | 609 | 121.6 | | |
| | `south-asia-peninsular` | <img src="regional/south-asia-peninsular/coverage.png" width="210"> | 671 | 122.9 | | |
| | `tibet` | <img src="regional/tibet/coverage.png" width="210"> | 1226 | 134.9 | | |
| ### North America | |
| | region | coverage | classes | fp32 MB | | |
| |---|---|---:|---:| | |
| | `canada-alaska` | <img src="regional/canada-alaska/coverage.png" width="210"> | 777 | 125.2 | | |
| | `north-america-east` | <img src="regional/north-america-east/coverage.png" width="210"> | 818 | 126.1 | | |
| | `north-america-west` | <img src="regional/north-america-west/coverage.png" width="210"> | 821 | 126.2 | | |
| ### South America | |
| | region | coverage | classes | fp32 MB | | |
| |---|---|---:|---:| | |
| | `amazonia` | <img src="regional/amazonia/coverage.png" width="210"> | 3182 | 177.0 | | |
| | `andes` | <img src="regional/andes/coverage.png" width="210"> | 3314 | 179.8 | | |
| | `eastern-brazil` | <img src="regional/eastern-brazil/coverage.png" width="210"> | 1972 | 150.9 | | |
| | `galapagos` | <img src="regional/galapagos/coverage.png" width="210"> | 156 | 111.9 | | |
| | `southern-cone` | <img src="regional/southern-cone/coverage.png" width="210"> | 1639 | 143.8 | | |
| ### Africa | |
| | region | coverage | classes | fp32 MB | | |
| |---|---|---:|---:| | |
| | `cape-verde` | <img src="regional/cape-verde/coverage.png" width="210"> | 162 | 112.0 | | |
| | `central-africa` | <img src="regional/central-africa/coverage.png" width="210"> | 1537 | 141.6 | | |
| | `east-africa` | <img src="regional/east-africa/coverage.png" width="210"> | 1824 | 147.8 | | |
| | `mauritius` | <img src="regional/mauritius/coverage.png" width="210"> | 90 | 110.4 | | |
| | `reunion` | <img src="regional/reunion/coverage.png" width="210"> | 92 | 110.5 | | |
| | `sao-tome-principe` | <img src="regional/sao-tome-principe/coverage.png" width="210"> | 133 | 111.4 | | |
| | `seychelles` | <img src="regional/seychelles/coverage.png" width="210"> | 137 | 111.4 | | |
| | `southern-africa` | <img src="regional/southern-africa/coverage.png" width="210"> | 820 | 126.1 | | |
| | `west-africa` | <img src="regional/west-africa/coverage.png" width="210"> | 1027 | 130.6 | | |
| ### Oceania | |
| | region | coverage | classes | fp32 MB | | |
| |---|---|---:|---:| | |
| | `australia-east` | <img src="regional/australia-east/coverage.png" width="210"> | 757 | 124.8 | | |
| | `hawaii` | <img src="regional/hawaii/coverage.png" width="210"> | 294 | 114.8 | | |
| | `new-caledonia` | <img src="regional/new-caledonia/coverage.png" width="210"> | 190 | 112.6 | | |
| | `new-zealand` | <img src="regional/new-zealand/coverage.png" width="210"> | 289 | 114.7 | | |
| ## Deployment | |
| - Point your host at a tile's model + its labels file (matching counts) and | |
| restart. Labels are `Scientific name_Common name`. | |
| - Confidence is a per-class sigmoid; slicing does not change a kept class's score | |
| (unlike a softmax model), so thresholds transfer directly from the full model. | |
| - Binaries are hosted here on HuggingFace; the derivation tooling lives separately. | |