Wav2vec2 Xlsr 300m Finnish Onnx

--- license: apache-2.0 language: - fi tags: - onnx - onnx-asr - wav2vec2 - automatic-speech-recognition base_model: - Finnish-NLP/wav2vec2-xlsr-300m-finnish-lm --- # wav2vec2-xlsr-300m-finnish-onnx (ONNX) ONNX export of [Finnish-NLP/wav2vec2-xlsr-300m-finnish-lm](https://huggingface.co/Finnish-NLP/wav2vec2-xlsr-300m-finnish-lm), a Finnish wav2vec2 CTC ASR model from Finnish-NLP, for use with [onnx-asr](https://github.com/istupakov/onnx-asr) (`wav2vec2-ctc` model type). Per-utterance zero-mean/unit-variance normalization is baked into the ONNX graph, masked by `input_lengths` for correct behavior with padded/batched input, so the model works with onnx-asr's plain `identity` preprocessor (raw 16kHz waveform in). ## Usage ```py import onnx_asr model = onnx_asr.load_model("OpenVoiceOS/wav2vec2-xlsr-300m-finnish-onnx") print(model.recognize("test.wav")) ``` ## Files * `model.onnx` / `model.onnx.data` — fp32 ONNX graph (inputs: `input_values` (batch, samples) float32, `input_lengths` (batch,) int64; output: `logprobs` (batch, frames, vocab) float32 log-softmax). * `vocab.txt` — CTC vocabulary in onnx-asr's `token id` format (word-delimiter -> `▁`, pad token -> `<blk>`). * `config.json` — `{"model_type": "wav2vec2-ctc", "subsampling_factor": 320}`. No int8 quantized variant is included yet -- `onnxruntime.quantization` does not currently support the `torch.onnx` dynamo-exported graph for this architecture. ## License Apache License 2.0, inherited from the base model.