Vctk Bsrnn Large Causal

--- tags: - espnet - audio - audio-to-audio language: en datasets: - VCTK_DEMAND license: cc-by-4.0 --- ## ESPnet2 ENH model ### `wyz/vctk_bsrnn_large_causal` This model was trained by wyz based on the universal_se_v1 recipe in [espnet](https://github.com/espnet/espnet/). More information can be found at https://github.com/Emrys365/se-scaling. ### Demo: How to use in ESPnet2 Follow the [ESPnet installation instructions](https://espnet.github.io/espnet/installation.html) if you haven't done that already. To use the model in the Python interface, you could use the following code: ```python import soundfile as sf from espnet2.bin.enh_inference import SeparateSpeech # For model downloading + loading model = SeparateSpeech.from_pretrained( model_tag="wyz/vctk_bsrnn_large_causal", normalize_output_wav=True, device="cuda", ) # For loading a downloaded model # model = SeparateSpeech( # train_config="exp_vctk/enh_train_enh_bsrnn_large_raw/config.yaml", # model_file="exp_vctk/enh_train_enh_bsrnn_large_raw/xxxx.pth", # normalize_output_wav=True, # device="cuda", # ) audio, fs = sf.read("/path/to/noisy/utt1.flac") enhanced = model(audio[None, :], fs=fs)[0] ``` <!-- Generated by ./scripts/utils/show_enh_score.sh --> # RESULTS ## Environments - date: `Wed Feb 28 12:15:02 EST 2024` - python version: `3.8.16 (default, Mar 2 2023, 03:21:46) [GCC 11.2.0]` - espnet version: `espnet 202304` - pytorch version: `pytorch 2.0.1+cu118` - Git hash: `443028662106472c60fe8bd892cb277e5b488651` - Commit date: `Thu May 11 03:32:59 2023 +0000` ## enhanced_test_16k |dataset|PESQ_WB|STOI|SAR|SDR|SIR|SI_SNR|OVRL|SIG|BAK|P808_MOS| |---|---|---|---|---|---|---|---|---|---|---| |chime4_et05_real_isolated_6ch_track|1.12|47.98|-3.67|-3.67|0.00|-31.29|2.43|2.85|3.39|3.19| |chime4_et05_simu_isolated_6ch_track|1.17|67.65|4.87|4.87|0.00|-1.00|2.30|2.65|3.47|2.87| |dns20_tt_synthetic_no_reverb|2.21|93.53|13.63|13.63|0.00|12.70|3.10|3.49|3.77|3.76| |reverb_et_real_8ch_multich|1.10|63.74|4.46|4.46|0.00|1.11|2.34|2.69|3.55|3.35| |reverb_et_simu_8ch_multich|1.58|82.53|8.76|8.76|0.00|-10.73|2.81|3.18|3.73|3.61| |whamr_tt_mix_single_reverb_max_16k|1.25|77.71|5.43|5.43|0.00|1.00|2.38|2.70|3.61|3.30| ## enhanced_test_48k |dataset|STOI|SAR|SDR|SIR|SI_SNR|OVRL|SIG|BAK|P808_MOS| |---|---|---|---|---|---|---|---|---|---| |vctk_noisy_tt_2spk|94.69|19.82|19.82|0.00|18.71|3.11|3.42|3.95|3.49| ##...

Source context: 2 downloads · 0 likes · Pipeline audio-to-audio · Library espnet · Repo wyz/vctk_bsrnn_large_causal