add model files

README.md

@@ -0,0 +1,80 @@
+---
+tags:
+- espnet
+- audio
+- automatic-speech-recognition
+- speech-translation
+language: multilingual
+datasets:
+- owsm_v3.1
+license: cc-by-4.0
+---
+
+## OWSM: Open Whisper-style Speech Model
+
+[OWSM](https://arxiv.org/abs/2309.13876) is an Open Whisper-style Speech Model from [CMU WAVLab](https://www.wavlab.org/). It reproduces Whisper-style training using publicly available data and an open-source toolkit [ESPnet](https://github.com/espnet/espnet).
+
+Our demo is available [here](https://huggingface.co/spaces/pyf98/OWSM_v3_demo).
+
+OWSM v3.1 is an improved version of OWSM v3. It significantly outperforms OWSM v3 in almost all evaluation benchmarks.
+We do not include any new training data. Instead, we utilize a state-of-the-art speech encoder, [E-Branchformer](https://arxiv.org/abs/2210.00077).
+
+This is a base size model which has 101M parameters and is trained on 180k hours of public speech data.
+Specifically, it supports the following speech-to-text tasks:
+- Speech recognition
+- Any-to-any-language speech translation
+- Utterance-level alignment
+- Long-form transcription
+- Language identification
+
+
+### Citing OWSM, Branchformers and ESPnet
+
+```BibTex
+@article{peng2023owsm,
+  title={Reproducing Whisper-Style Training Using an Open-Source Toolkit and Publicly Available Data},
+  author={Yifan Peng and Jinchuan Tian and Brian Yan and Dan Berrebbi and Xuankai Chang and Xinjian Li and Jiatong Shi and Siddhant Arora and William Chen and Roshan Sharma and Wangyou Zhang and Yui Sudo and Muhammad Shakeel and Jee-weon Jung and Soumi Maiti and Shinji Watanabe},
+  journal={arXiv preprint arXiv:2309.13876},
+  year={2023}
+}
+@inproceedings{peng23b_interspeech,
+  author={Yifan Peng and Kwangyoun Kim and Felix Wu and Brian Yan and Siddhant Arora and William Chen and Jiyang Tang and Suwon Shon and Prashant Sridhar and Shinji Watanabe},
+  title={{A Comparative Study on E-Branchformer vs Conformer in Speech Recognition, Translation, and Understanding Tasks}},
+  year=2023,
+  booktitle={Proc. INTERSPEECH 2023},
+  pages={2208--2212},
+  doi={10.21437/Interspeech.2023-1194}
+}
+@inproceedings{kim2023branchformer,
+  title={E-branchformer: Branchformer with enhanced merging for speech recognition},
+  author={Kim, Kwangyoun and Wu, Felix and Peng, Yifan and Pan, Jing and Sridhar, Prashant and Han, Kyu J and Watanabe, Shinji},
+  booktitle={2022 IEEE Spoken Language Technology Workshop (SLT)},
+  pages={84--91},
+  year={2023},
+  organization={IEEE}
+}
+@InProceedings{pmlr-v162-peng22a,
+  title = 	 {Branchformer: Parallel {MLP}-Attention Architectures to Capture Local and Global Context for Speech Recognition and Understanding},
+  author =       {Peng, Yifan and Dalmia, Siddharth and Lane, Ian and Watanabe, Shinji},
+  booktitle = 	 {Proceedings of the 39th International Conference on Machine Learning},
+  pages = 	 {17627--17643},
+  year = 	 {2022},
+  editor = 	 {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan},
+  volume = 	 {162},
+  series = 	 {Proceedings of Machine Learning Research},
+  month = 	 {17--23 Jul},
+  publisher =    {PMLR},
+  pdf = 	 {https://proceedings.mlr.press/v162/peng22a/peng22a.pdf},
+  url = 	 {https://proceedings.mlr.press/v162/peng22a.html},
+  abstract = 	 {Conformer has proven to be effective in many speech processing tasks. It combines the benefits of extracting local dependencies using convolutions and global dependencies using self-attention. Inspired by this, we propose a more flexible, interpretable and customizable encoder alternative, Branchformer, with parallel branches for modeling various ranged dependencies in end-to-end speech processing. In each encoder layer, one branch employs self-attention or its variant to capture long-range dependencies, while the other branch utilizes an MLP module with convolutional gating (cgMLP) to extract local relationships. We conduct experiments on several speech recognition and spoken language understanding benchmarks. Results show that our model outperforms both Transformer and cgMLP. It also matches with or outperforms state-of-the-art results achieved by Conformer. Furthermore, we show various strategies to reduce computation thanks to the two-branch architecture, including the ability to have variable inference complexity in a single trained model. The weights learned for merging branches indicate how local and global dependencies are utilized in different layers, which benefits model designing.}
+}
+@inproceedings{watanabe2018espnet,
+  author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Yalta and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai},
+  title={{ESPnet}: End-to-End Speech Processing Toolkit},
+  year={2018},
+  booktitle={Proceedings of Interspeech},
+  pages={2207--2211},
+  doi={10.21437/Interspeech.2018-1456},
+  url={http://dx.doi.org/10.21437/Interspeech.2018-1456}
+}
+```

data/token_list/bpe_unigram50000/bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d6327da127e870bcb8c737dceb3bd47ccbce63da74ddb094f64afe313d68c8c
+size 1041297

exp/s2t_stats_raw_bpe50000/train/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ef4b5e465110edf32eec024cf2427eedd677f5733bb87d6b2131e6984a6e13f
+size 1402

exp/s2t_train_s2t_ebf_conv2d_size384_e6_d6_piecewise_lr1e-3_warmup60k_flashattn_lessreg_raw_bpe50000/RESULTS.md

@@ -0,0 +1,9 @@
+<!-- Generated by scripts/utils/show_asr_result.sh -->
+# RESULTS
+## Environments
+- date: `Thu Jan 18 13:29:51 CST 2024`
+- python version: `3.10.10 (main, Mar 21 2023, 18:45:11) [GCC 11.2.0]`
+- espnet version: `espnet 202308`
+- pytorch version: `pytorch 1.13.1`
+- Git hash: `e5c058e28bd8db071b19cb45688165b7013c0938`
+  - Commit date: `Tue Dec 26 21:55:42 2023 -0600`

exp/s2t_train_s2t_ebf_conv2d_size384_e6_d6_piecewise_lr1e-3_warmup60k_flashattn_lessreg_raw_bpe50000/valid.total_count.ave_5best.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99e5de1865e2c98308b41ce6f28b7f658bec7b274da60f37b219a99279d43f3a
+size 404971245

meta.yaml

@@ -0,0 +1,8 @@
+espnet: '202308'
+files:
+  s2t_model_file: exp/s2t_train_s2t_ebf_conv2d_size384_e6_d6_piecewise_lr1e-3_warmup60k_flashattn_lessreg_raw_bpe50000/valid.total_count.ave_5best.pth
+python: 3.10.10 (main, Mar 21 2023, 18:45:11) [GCC 11.2.0]
+timestamp: 1705960184.857766
+torch: 1.13.1
+yaml_files:
+  s2t_train_config: exp/s2t_train_s2t_ebf_conv2d_size384_e6_d6_piecewise_lr1e-3_warmup60k_flashattn_lessreg_raw_bpe50000/config.yaml