add files

README.md

@@ -0,0 +1,120 @@
+---
+language:
+- ca
+- en
+datasets:
+- covost2
+tags:
+- audio
+- speech-translation
+- automatic-speech-recognition
+license: MIT
+---
+
+
+# S2T-SMALL-COVOST2-CA-EN-ST
+
+`s2t-small-covost2-ca-en-st` is a Speech to Text Transformer (S2T) model trained for end-to-end Speech Translation (ST).
+The S2T model was proposed in [this paper](https://arxiv.org/abs/2010.05171) and released in
+[this repository](https://github.com/pytorch/fairseq/tree/master/examples/speech_to_text)
+
+
+## Model description
+
+S2T is a transformer-based seq2seq (encoder-decoder) model designed for end-to-end Automatic Speech Recognition (ASR) and Speech
+Translation (ST). It uses a convolutional downsampler to reduce the length of speech inputs by 3/4th before they are
+fed into the encoder. The model is trained with standard autoregressive cross-entropy loss and generates the
+transcripts/translations autoregressively.
+
+## Intended uses & limitations
+
+This model can be used for end-to-end Catalan speech to English text translation.
+See the [model hub](https://huggingface.co/models?filter=speech_to_text) to look for other S2T checkpoints.
+
+
+### How to use
+
+As this a standard sequence to sequence transformer model, you can use the `generate` method to generate the
+transcripts by passing the speech features to the model.
+
+*Note: The `Speech2TextProcessor` object uses [torchaudio](https://github.com/pytorch/audio)  to extract the
+filter bank features. Make sure to install the `torchaudio` package before running this example.*
+
+You could either install those as extra speech dependancies with
+`pip install transformers"[speech, sentencepiece]"` or install the packages seperatly 
+with `pip install torchaudio sentencepiece`.
+
+
+```python
+import torch
+from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
+from datasets import load_dataset
+import soundfile as sf
+
+model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-covost2-ca-en-st")
+processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-covost2-ca-en-st")
+
+def map_to_array(batch):
+    speech, _ = sf.read(batch["file"])
+    batch["speech"] = speech
+    return batch
+
+ds = load_dataset(
+    "patrickvonplaten/librispeech_asr_dummy",
+    "clean",
+    split="validation"
+)
+ds = ds.map(map_to_array)
+
+inputs = processor(
+    ds["speech"][0],
+    sampling_rate=48_000,
+    return_tensors="pt"
+)
+generated_ids = model.generate(input_ids=inputs["input_features"], attention_mask=inputs["attention_mask"])
+
+translation = processor.batch_decode(generated_ids, skip_special_tokens=True)
+```
+
+
+## Training data
+
+The s2t-small-covost2-ca-en-st is trained on Catalan-English subset of [CoVoST2](https://github.com/facebookresearch/covost).
+CoVoST is a large-scale multilingual ST corpus based on [Common Voice](https://arxiv.org/abs/1912.06670), created to to foster 
+ST research with the largest ever open dataset
+
+
+## Training procedure
+
+### Preprocessing
+
+The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from
+WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization)
+is applied to each example.
+
+The texts are lowercased and tokenized using character based SentencePiece vocab.
+
+
+### Training
+
+The model is trained with standard autoregressive cross-entropy loss and using [SpecAugment](https://arxiv.org/abs/1904.08779).
+The encoder receives speech features, and the decoder generates the transcripts autoregressively. To accelerate
+model training and for better performance the encoder is pre-trained for English ASR.
+
+## Evaluation results
+
+CoVOST2 test results for ca-en (BLEU score): 17.85
+
+
+
+### BibTeX entry and citation info
+
+```bibtex
+@inproceedings{wang2020fairseqs2t,
+  title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq},
+  author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino},
+  booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations},
+  year = {2020},
+}
+
+```

config.json

@@ -0,0 +1,46 @@
+{
+  "activation_dropout": 0.1,
+  "activation_function": "relu",
+  "architectures": [
+    "Speech2TextForConditionalGeneration"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": 0.0,
+  "conv_channels": 1024,
+  "conv_kernel_sizes": [
+    5,
+    5
+  ],
+  "d_model": 256,
+  "decoder_attention_heads": 4,
+  "decoder_ffn_dim": 2048,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 6,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "early_stopping": true,
+  "encoder_attention_heads": 4,
+  "encoder_ffn_dim": 2048,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "init_std": 0.02,
+  "input_channels": 1,
+  "input_feat_per_channel": 80,
+  "is_encoder_decoder": true,
+  "max_length": 200,
+  "max_source_positions": 6000,
+  "max_target_positions": 1024,
+  "model_type": "speech_to_text",
+  "num_beams": 5,
+  "num_conv_layers": 2,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "tie_word_embeddings": false,
+  "transformers_version": "4.4.0.dev0",
+  "use_cache": true,
+  "vocab_size": 132
+}

preprocessor_config.json

@@ -0,0 +1,11 @@
+{
+  "do_ceptral_normalize": true,
+  "feature_size": 80,
+  "normalize_means": true,
+  "normalize_vars": true,
+  "num_mel_bins": 80,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": true,
+  "sampling_rate": 48000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38721402333fe35888f09a0c1b45dd3005d099fc12e96f5f0060786257a02497
+size 108297733

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba4c9a357b3d07caf9fb50b157740f7b5431a20612a46672646a35bb0e49aa54
+size 239098

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "do_upper_case": false, "do_lower_case": false, "tgt_lang": null, "lang_codes": null, "tokenizer_file": null}

vocab.json

@@ -0,0 +1,134 @@
+{
+  "<s>": 0,
+  "<pad>": 1,
+  "</s>": 2,
+  "<unk>": 3,
+  "\u2581": 4,
+  "e": 5,
+  "t": 6,
+  "a": 7,
+  "o": 8,
+  "i": 9,
+  "n": 10,
+  "s": 11,
+  "r": 12,
+  "h": 13,
+  "l": 14,
+  "d": 15,
+  "c": 16,
+  "u": 17,
+  "m": 18,
+  "f": 19,
+  ".": 20,
+  "p": 21,
+  "g": 22,
+  "y": 23,
+  "w": 24,
+  "b": 25,
+  "v": 26,
+  ",": 27,
+  "T": 28,
+  "k": 29,
+  "I": 30,
+  "A": 31,
+  "W": 32,
+  "S": 33,
+  "x": 34,
+  "\u2019": 35,
+  "C": 36,
+  "H": 37,
+  "M": 38,
+  "L": 39,
+  "B": 40,
+  "P": 41,
+  "F": 42,
+  "q": 43,
+  "z": 44,
+  "'": 45,
+  "j": 46,
+  "-": 47,
+  "E": 48,
+  "O": 49,
+  "D": 50,
+  "R": 51,
+  "G": 52,
+  "N": 53,
+  "V": 54,
+  "?": 55,
+  "J": 56,
+  "U": 57,
+  "!": 58,
+  "\u00e0": 59,
+  "Y": 60,
+  ":": 61,
+  "\u00f3": 62,
+  ";": 63,
+  "\u00ed": 64,
+  "\u00e7": 65,
+  "K": 66,
+  "0": 67,
+  "1": 68,
+  "\u00e8": 69,
+  "\u00e9": 70,
+  "2": 71,
+  "Q": 72,
+  "\"": 73,
+  "X": 74,
+  "\u00fa": 75,
+  "\u201d": 76,
+  "\u201c": 77,
+  "(": 78,
+  ")": 79,
+  "9": 80,
+  "\u00f2": 81,
+  "5": 82,
+  "\u2014": 83,
+  "Z": 84,
+  "6": 85,
+  "3": 86,
+  "7": 87,
+  "8": 88,
+  "\u2018": 89,
+  "4": 90,
+  "\u00e1": 91,
+  "\u00fc": 92,
+  "[": 93,
+  "]": 94,
+  "\u00c0": 95,
+  "/": 96,
+  "\u00ef": 97,
+  "\u00d2": 98,
+  "\u00f1": 99,
+  "\u00ab": 100,
+  "\u00bb": 101,
+  "%": 102,
+  "\u2013": 103,
+  "\u00b7": 104,
+  "\u00c9": 105,
+  "\u00e2": 106,
+  "\u00f4": 107,
+  "\u00c8": 108,
+  "\u00c1": 109,
+  "\u00da": 110,
+  "\u0301": 111,
+  "@": 112,
+  "`": 113,
+  "\u00f8": 114,
+  "\u20ac": 115,
+  "_": 116,
+  "\u016b": 117,
+  "+": 118,
+  "=": 119,
+  "\\": 120,
+  "\u00b0": 121,
+  "\u00cd": 122,
+  "\u00e3": 123,
+  "\u00ea": 124,
+  "\u00ec": 125,
+  "\u0107": 126,
+  "\u0117": 127,
+  "\u0142": 128,
+  "\u0151": 129,
+  "\u017d": 130,
+  "\u017e": 131
+}