first

.gitattributes

@@ -25,3 +25,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.txt filter=lfs diff=lfs merge=lfs -text
+*unigram*.* filter=lfs diff=lfs merge=lfs -text

add_kenlm.py

@@ -0,0 +1,34 @@
+import argparse
+from transformers import AutoProcessor
+from transformers import Wav2Vec2ProcessorWithLM
+from pyctcdecode import build_ctcdecoder
+
+
+def main(args):
+    processor = AutoProcessor.from_pretrained(args.model_name_or_path)
+    vocab_dict = processor.tokenizer.get_vocab()
+    sorted_vocab_dict = {
+        k.lower(): v for k, v in sorted(vocab_dict.items(), key=lambda item: item[1])
+    }
+    decoder = build_ctcdecoder(
+        labels=list(sorted_vocab_dict.keys()),
+        kenlm_model_path=args.kenlm_model_path,
+    )
+    processor_with_lm = Wav2Vec2ProcessorWithLM(
+        feature_extractor=processor.feature_extractor,
+        tokenizer=processor.tokenizer,
+        decoder=decoder,
+    )
+    processor_with_lm.save_pretrained(args.model_name_or_path)
+    print(f"Run: ~/bin/build_binary language_model/*.arpa language_model/5gram.bin -T $(pwd) && rm language_model/*.arpa")
+
+def parse_args():
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--model_name_or_path', default="./", help='Model name or path. Defaults to ./')
+    parser.add_argument('--kenlm_model_path', required=True, help='Path to KenLM arpa file.')
+    args = parser.parse_args()
+    return args
+
+if __name__ == "__main__":
+    args = parse_args()
+    main(args)

added_tokens.json

@@ -0,0 +1 @@
+{"<s>": 32, "</s>": 33}

alphabet.json

@@ -0,0 +1 @@
+{"labels": [" ", "a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z", "\u00e5", "\u00e6", "\u00f8", "\u2047", "", "<s>", "</s>"], "is_bpe": false}

config.json

@@ -0,0 +1,107 @@
+{
+  "_name_or_path": "./",
+  "activation_dropout": 0.055,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.094,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 1024,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": true,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": true,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": true,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.04,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.047,
+  "hidden_size": 1280,
+  "initializer_range": 0.02,
+  "intermediate_size": 5120,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.041,
+  "mask_feature_length": 64,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.25,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.082,
+  "model_type": "wav2vec2",
+  "num_adapter_layers": 3,
+  "num_attention_heads": 16,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 48,
+  "num_negatives": 100,
+  "output_hidden_size": 1280,
+  "pad_token_id": 31,
+  "proj_codevector_dim": 1024,
+  "tdnn_dilation": [
+    1,
+    2,
+    3,
+    1,
+    1
+  ],
+  "tdnn_dim": [
+    512,
+    512,
+    512,
+    512,
+    1500
+  ],
+  "tdnn_kernel": [
+    5,
+    3,
+    3,
+    1,
+    1
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.17.0.dev0",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 34,
+  "xvector_output_dim": 512
+}

eval.py

@@ -0,0 +1,151 @@
+#!/usr/bin/env python3
+import argparse
+import re
+from typing import Dict
+
+import torch
+from datasets import Audio, Dataset, load_dataset, load_metric
+
+from transformers import AutoFeatureExtractor, pipeline
+
+
+def log_results(result: Dataset, args: Dict[str, str]):
+    """DO NOT CHANGE. This function computes and logs the result metrics."""
+
+    log_outputs = args.log_outputs
+    dataset_id = "_".join(args.dataset.split("/") + [args.config, args.split])
+
+    # load metric
+    wer = load_metric("wer")
+    cer = load_metric("cer")
+
+    # compute metrics
+    wer_result = wer.compute(references=result["target"], predictions=result["prediction"])
+    cer_result = cer.compute(references=result["target"], predictions=result["prediction"])
+
+    # print & log results
+    result_str = f"WER: {wer_result}\n" f"CER: {cer_result}"
+    print(result_str)
+
+    with open(f"{dataset_id}_eval_results.txt", "w") as f:
+        f.write(result_str)
+
+    # log all results in text file. Possibly interesting for analysis
+    if log_outputs is not None:
+        pred_file = f"log_{dataset_id}_predictions.txt"
+        target_file = f"log_{dataset_id}_targets.txt"
+
+        with open(pred_file, "w") as p, open(target_file, "w") as t:
+
+            # mapping function to write output
+            def write_to_file(batch, i):
+                p.write(f"{i}" + "\n")
+                p.write(batch["prediction"] + "\n")
+                t.write(f"{i}" + "\n")
+                t.write(batch["target"] + "\n")
+
+            result.map(write_to_file, with_indices=True)
+
+
+def normalize_text(text: str) -> str:
+    """DO ADAPT FOR YOUR USE CASE. this function normalizes the target text."""
+
+    chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“\%\‘\”\�\'\–\_\\\+\#\/]'  # noqa: W605 IMPORTANT: this should correspond to the chars that were ignored during training
+
+    text = re.sub(chars_to_ignore_regex, "", text.lower()) + " "
+    text = re.sub('[áàâ]', 'a', text)
+    text = re.sub('[ä]', 'æ', text)
+    text = re.sub('[éèëê]', 'e', text)
+    text = re.sub('[íìïî]', 'i', text)
+    text = re.sub('[óòöô]', 'o', text)
+    text = re.sub('[ö]', 'ø', text)
+    text = re.sub('[ç]', 'c', text)
+    text = re.sub('[úùüû]', 'u', text)
+    text = re.sub('\s', ' ', text)
+    text = re.sub('<ee>', 'eee', text)
+    text = re.sub('<qq>', 'qqq', text)
+    text = re.sub('<mm>', 'mmm', text)
+    text = re.sub('<inaudible>', 'xxx', text)
+    text = re.sub('[<>]', '', text)
+
+    # # In addition, we can normalize the target text, e.g. removing new lines characters etc...
+    # # note that order is important here!
+    # token_sequences_to_ignore = ["\n\n", "\n", "   ", "  "]
+
+    # for t in token_sequences_to_ignore:
+    #     text = " ".join(text.split(t))
+
+    return text
+
+
+def main(args):
+    # load dataset
+    dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
+
+    # for testing: only process the first two examples as a test
+    # dataset = dataset.select(range(10))
+
+    # load processor
+    feature_extractor = AutoFeatureExtractor.from_pretrained(args.model_id)
+    sampling_rate = feature_extractor.sampling_rate
+
+    # resample audio
+    dataset = dataset.cast_column("audio", Audio(sampling_rate=sampling_rate))
+
+    # load eval pipeline
+    if args.device is None:
+        args.device = 0 if torch.cuda.is_available() else -1
+    asr = pipeline("automatic-speech-recognition", model=args.model_id, device=args.device)
+
+    # map function to decode audio
+    def map_to_pred(batch):
+        prediction = asr(
+            batch["audio"]["array"], chunk_length_s=args.chunk_length_s, stride_length_s=args.stride_length_s
+        )
+
+        batch["prediction"] = prediction["text"]
+        batch["target"] = normalize_text(batch["text"])
+        return batch
+
+    # run inference on all examples
+    result = dataset.map(map_to_pred, remove_columns=dataset.column_names)
+
+    # compute and log_results
+    # do not change function below
+    log_results(result, args)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument(
+        "--model_id", type=str, required=True, help="Model identifier. Should be loadable with 🤗 Transformers"
+    )
+    parser.add_argument(
+        "--dataset",
+        type=str,
+        required=True,
+        help="Dataset name to evaluate the `model_id`. Should be loadable with 🤗 Datasets",
+    )
+    parser.add_argument(
+        "--config", type=str, required=True, help="Config of the dataset. *E.g.* `'en'`  for Common Voice"
+    )
+    parser.add_argument("--split", type=str, required=True, help="Split of the dataset. *E.g.* `'test'`")
+    parser.add_argument(
+        "--chunk_length_s", type=float, default=None, help="Chunk length in seconds. Defaults to 5 seconds."
+    )
+    parser.add_argument(
+        "--stride_length_s", type=float, default=None, help="Stride of the audio chunks. Defaults to 1 second."
+    )
+    parser.add_argument(
+        "--log_outputs", action="store_true", help="If defined, write outputs to log file for analysis."
+    )
+    parser.add_argument(
+        "--device",
+        type=int,
+        default=None,
+        help="The device to run the pipeline on. -1 for CPU (default), 0 for the first GPU and so on.",
+    )
+    args = parser.parse_args()
+
+    main(args)

grid.csv

@@ -0,0 +1,101 @@
+alpha,beta,wer,cer
+0.001,0.001,0.07689050968468251,0.026711121521565424
+0.001,0.01,0.07686164326883294,0.026699492822600875
+0.001,0.1,0.07693862037776517,0.026702815308019316
+0.001,0.25,0.07732350592242632,0.026749330103877506
+0.001,0.5,0.07791045637803458,0.026812457326827906
+0.001,0.75,0.07878607099213872,0.02695532419982092
+0.001,1,0.07969055202209244,0.02708490113114016
+0.001,1.5,0.08199986529005937,0.02737894109067229
+0.001,2,0.08457859843928911,0.027772655612757678
+0.001,3,0.09057319079738663,0.028741160112233557
+0.01,0.001,0.07580320802101476,0.026571577133990852
+0.01,0.01,0.07583207443686434,0.026571577133990852
+0.01,0.1,0.07596678437749574,0.026606463230884496
+0.01,0.25,0.07640940275385608,0.026649655541324244
+0.01,0.5,0.07688088754606599,0.02667623542467178
+0.01,0.75,0.07770839146708748,0.02679916738515414
+0.01,1,0.0786513610515073,0.026963630413367023
+0.01,1.5,0.08097029645809077,0.027274282799991362
+0.01,2,0.08339507538945606,0.027613176312672457
+0.01,3,0.08929344636138828,0.02853018228816248
+0.1,0.001,0.0702319897620445,0.025917047506557756
+0.1,0.01,0.07021274548481145,0.025907080050302428
+0.1,0.1,0.07033783328682633,0.025920369991976198
+0.1,0.25,0.07057838675223956,0.025960239816997504
+0.1,0.5,0.0711268486533817,0.026010077098274135
+0.1,0.75,0.07168493269314037,0.026078188049352198
+0.1,1,0.07225263887151559,0.026169556398359356
+0.1,1.5,0.07360936041644615,0.026289165873423274
+0.1,2,0.07566849808038334,0.0265516422214802
+0.1,3,0.08056616663619656,0.027264315343736033
+0.25,0.001,0.06741270314740154,0.025528316712600026
+0.25,0.01,0.06744156956325113,0.025539945411564575
+0.25,0.1,0.06741270314740154,0.025521671741763143
+0.25,0.25,0.06753779094941642,0.025523332984472364
+0.25,0.5,0.06765325661281477,0.025533300440727692
+0.25,0.75,0.06790343221684451,0.02557649275116744
+0.25,1,0.06817285209810732,0.025568186537621333
+0.25,1.5,0.06890413463296352,0.025651248673082386
+0.25,2,0.0696642835836693,0.025696102226231355
+0.25,3,0.07246432592107922,0.02611141290353662
+0.5,0.001,0.06591164952322304,0.025593105178259648
+0.5,0.01,0.06596938235492221,0.025601411391805756
+0.5,0.1,0.06597900449353873,0.02559808890638731
+0.5,0.25,0.06613295871140319,0.02557649275116744
+0.5,0.5,0.06615220298863625,0.025506720557380153
+0.5,0.75,0.06619069154310236,0.0254867856448695
+0.5,1,0.06632540148373378,0.025505059314670932
+0.5,1.5,0.0667006648897784,0.025520010499053922
+0.5,2,0.06719139395922137,0.025490108130287945
+0.5,3,0.06871169186063295,0.02567118358559304
+0.75,0.001,0.06652746639468088,0.025825679157550598
+0.75,0.01,0.06651784425606436,0.025825679157550598
+0.75,0.1,0.06646011142436518,0.02581737294400449
+0.75,0.25,0.0663350236223503,0.025792454303366175
+0.75,0.5,0.06650822211744782,0.025809066730458387
+0.75,0.75,0.06641200073128253,0.025799099274203062
+0.75,1,0.0663638900381999,0.025755906963763314
+0.75,1.5,0.06648897784021476,0.025729327080415774
+0.75,2,0.06680650841456022,0.025714375896032787
+0.75,3,0.0673645924543189,0.025641281216827058
+1,0.001,0.06927939803900815,0.02653502979438799
+1,0.01,0.06926015376177509,0.02653336855167877
+1,0.1,0.06918317665284286,0.026548319736061757
+1,0.25,0.06900997815774534,0.02645861262976382
+1,0.5,0.068730936137866,0.026368905523465883
+1,0.75,0.06865395902893377,0.026353954339082896
+1,1,0.06855773764276848,0.02630411705780626
+1,1.5,0.06833642845458832,0.02620776498067144
+1,2,0.06804776429609245,0.026071543078515315
+1,3,0.06796116504854369,0.025993464671181923
+1.5,0.001,0.07664033408065277,0.027842427806544966
+1.5,0.01,0.07664995621926929,0.02785073402009107
+1.5,0.1,0.07654411269448748,0.02783910532112652
+1.5,0.25,0.07627469281322467,0.027794251767977552
+1.5,0.5,0.07588980726856351,0.027689593477296628
+1.5,0.75,0.07547605530805276,0.027579951458488038
+1.5,1,0.07511041404062467,0.02752014672095608
+1.5,1.5,0.07450421930778335,0.027398876003182943
+1.5,2,0.07381142532739327,0.027354022450033974
+1.5,3,0.07257016944586103,0.027058321247792623
+2,0.001,0.08493461756810068,0.02921793676978
+2,0.01,0.08491537329086762,0.02921627552707078
+2,0.1,0.08472293051853705,0.029146503333283495
+2,0.25,0.08439577780557507,0.02908835983846076
+2,0.5,0.08399164798368085,0.029020248887382697
+2,0.75,0.08366449527071887,0.028968750363396842
+2,1,0.0831737662012759,0.02886741455813436
+2,1.5,0.08208646453760812,0.028692984073666147
+2,2,0.0808452086560759,0.028500279919396503
+2,3,0.07907473515063458,0.028187966290062944
+3,0.001,0.09346945452096182,0.03065657295596544
+3,0.01,0.0934598323823453,0.03065823419867466
+3,0.1,0.09336361099618001,0.030631654315327123
+3,0.25,0.09322890105554861,0.030629993072617902
+3,0.5,0.0928632597881205,0.03056188212153984
+3,0.75,0.09247837424345935,0.030490448685043334
+3,1,0.0922955536097453,0.030488787442334114
+3,1.5,0.09158351535212217,0.030339275598504217
+3,2,0.09083298854003291,0.030241262278660176
+3,3,0.08947626699510233,0.029978785930603248

grid.py

@@ -0,0 +1,34 @@
+#!/usr/bin/env python
+from argparse import ArgumentDefaultsHelpFormatter
+from collections import namedtuple
+from functools import partialmethod
+import json
+
+from tqdm import tqdm
+from eval import main
+
+
+# tqdm.__init__ = partialmethod(tqdm.__init__, disable=True)
+
+Args = namedtuple("Args", "model_id dataset config split log_outputs chunk_length_s stride_length_s device")
+args = Args("./", "NbAiLab/NPSC", "16K_mp3_bokmaal", "test", True, None, None, 0)
+with open("grid.csv", "w") as grid:
+    grid.write("alpha,beta,wer,cer")
+    for alpha in [0.001, 0.01, 0.1, 0.25, 0.5, 0.75, 1, 1.5, 2, 3]:
+        for beta in [0.001, 0.01, 0.1, 0.25, 0.5, 0.75, 1, 1.5, 2, 3]:
+            with open("./language_model/attrs.json", "r") as attrs_file:
+                attrs = json.load(attrs_file)
+            attrs["alpha"] = alpha
+            attrs["beta"] = beta
+            with open("./language_model/attrs.json", "w") as attrs_file:
+                json.dump(attrs, attrs_file)
+            print(f"alpha = {alpha}, beta = {beta}")
+            main(args)
+            with open("NbAiLab_NPSC_16K_mp3_bokmaal_test_eval_results.txt") as results_file:
+                results = results_file.read().strip().split("\n")
+            wer = float(results[0][5:])
+            cer = float(results[1][5:])
+            grid.write(f"\n{alpha},{beta},{wer},{cer}")
+            print("--------------")
+
+

language_model/5gram.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b41c24c63f2f0585bea83666369593f3b3e6d047f327a90f36ebca2c35ef0ff
+size 4243671427

language_model/attrs.json

@@ -0,0 +1 @@
+{"alpha": 0.5, "beta": 0.0, "unk_score_offset": -10.0, "score_boundary": true}

language_model/unigrams.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac3e71ca49838ca355df6fdcb8d89344a5a9bf9e1a76587cdf5df1367c19b9a9
+size 16759269

log_NbAiLab_NPSC_16K_mp3_bokmaal_test_eval_results.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:658588cf79099ab697bc9ff0c5d53fe956e3c9fdf385b4715129d4148be99495
+size 49

log_NbAiLab_NPSC_16K_mp3_bokmaal_test_eval_results_no_lang_model.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcb02a32f3393edc8aaa2ec5dcb5c2f4d8c7a1471d4a2d553dd85fed00579e54
+size 50

log_NbAiLab_NPSC_16K_mp3_bokmaal_test_predictions.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ef742d63f7ba1dffc7c27f02beff74b79df5b904e3ba56f4df7c0c41bb6719e
+size 648798

log_NbAiLab_NPSC_16K_mp3_bokmaal_test_predictions_no_lang_model.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:087bb268470c79b3b5e102928cbce002bfcd6ec791db8b0c0600a775bcc7a90f
+size 649987

log_NbAiLab_NPSC_16K_mp3_bokmaal_test_targets.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7e27a71ec76c0ded0b0cf852b9d42c620e1c5863bcf9a06fcbbbbd924e9d1b9
+size 654671

log_NbAiLab_NPSC_16K_mp3_bokmaal_test_targets_no_lang_model.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7e27a71ec76c0ded0b0cf852b9d42c620e1c5863bcf9a06fcbbbbd924e9d1b9
+size 654671

preprocessor_config.json

@@ -0,0 +1,10 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0,
+  "processor_class": "Wav2Vec2ProcessorWithLM",
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01edd44641353c2f34c79eb346e6153816c4f35f49334bba6a1a108cb99f9ca5
+size 3850486961

run.sh

@@ -0,0 +1,40 @@
+python run_speech_recognition_ctc.py \
+        --dataset_name="NbAiLab/NPSC" \
+        --model_name_or_path="facebook/wav2vec2-xls-r-1b" \
+        --dataset_config_name="16K_mp3_bokmaal" \
+        --output_dir="./" \
+        --overwrite_output_dir \
+        --num_train_epochs="40" \
+        --per_device_train_batch_size="12" \
+        --per_device_eval_batch_size="12" \
+        --gradient_accumulation_steps="2" \
+        --learning_rate="2e-5" \
+        --warmup_steps="2000" \
+        --length_column_name="input_length" \
+        --evaluation_strategy="steps" \
+        --text_column_name="text" \
+        --save_steps="500" \
+        --eval_steps="500" \
+        --logging_steps="100" \
+        --layerdrop="0.041" \
+        --attention_dropout="0.094" \
+        --activation_dropout="0.055" \
+        --hidden_dropout="0.047" \
+        --save_total_limit="3" \
+        --freeze_feature_encoder \
+        --feat_proj_dropout="0.04" \
+        --mask_time_prob="0.082" \
+        --mask_time_length="10" \
+        --mask_feature_prob="0.25" \
+        --mask_feature_length="64" \
+        --gradient_checkpointing \
+        --min_duration_in_seconds="0.5" \
+        --max_duration_in_seconds="30.0" \
+	--ctc_zero_infinity=True \
+        --use_auth_token \
+        --seed="42" \
+        --fp16 \
+        --group_by_length \
+        --do_train --do_eval \
+        --push_to_hub \
+        --preprocessing_num_workers="16"

runs/Feb06_21-00-20_job-c93f32d8-97c5-48e7-b5ec-c6c950f627ba/1644181327.7862043/events.out.tfevents.1644181327.job-c93f32d8-97c5-48e7-b5ec-c6c950f627ba.1233455.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e42ad114d53443aa03f4085977f0278259c263c0e55643cbaa71e38f85416ac1
+size 4776

runs/Feb06_21-00-20_job-c93f32d8-97c5-48e7-b5ec-c6c950f627ba/events.out.tfevents.1644181327.job-c93f32d8-97c5-48e7-b5ec-c6c950f627ba.1233455.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0ebdea7a8a0435aad29811dd38d7788869adc2fe05cfc79439e65e4294566da
+size 4716

runs/Feb06_21-06-35_job-c93f32d8-97c5-48e7-b5ec-c6c950f627ba/1644181697.767686/events.out.tfevents.1644181697.job-c93f32d8-97c5-48e7-b5ec-c6c950f627ba.1235981.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce7b1a25c6e7bbf6d2ed9adbdc66ff7df07b371ba42014692b7c814df923ebb2
+size 4776

runs/Feb06_21-06-35_job-c93f32d8-97c5-48e7-b5ec-c6c950f627ba/events.out.tfevents.1644181697.job-c93f32d8-97c5-48e7-b5ec-c6c950f627ba.1235981.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe271cf8605594bd4a0e1505b702d0fc514760349acaf2eb5e7c2e3b80cf52db
+size 35597

runs/Feb07_20-59-13_dante/1644264797.365929/events.out.tfevents.1644264797.dante.3536313.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ffaa426d205ad6d41b30a030b281139364685c886039e88dcc284cb2a3a0f617
+size 4741

runs/Feb07_20-59-13_dante/events.out.tfevents.1644264797.dante.3536313.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5df3d4e4263a791a1b8a910fc20582aab5f452f23785aabf3b0127d3241bef44
+size 7325

runs/Feb07_22-48-25_dante/1644270591.175515/events.out.tfevents.1644270591.dante.3603912.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43d94d5a4ce2f9a3c0470505f7982a3d6ed9f0edf871889542ad0492f815bb19
+size 4741

runs/Feb07_22-48-25_dante/events.out.tfevents.1644270591.dante.3603912.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf808b11fe3fbb5ab86b38a08a766a78cb23b58d81fc7c716fd0768ed64b102b
+size 162164

runs/Feb07_22-48-25_dante/events.out.tfevents.1644597285.dante.3603912.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02ca0e871dca33ce2c29f8fcff4c737976489a20f3935f2766148a74c94ca749
+size 364

runs/Feb11_17-40-48_dante/1644598421.1311183/events.out.tfevents.1644598421.dante.2344752.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25ed8c04ecf820de1e43849ef6d73fd1d2bfc065b4b8e0ea55d4e4e4acc4b54a
+size 4741

runs/Feb11_17-40-48_dante/events.out.tfevents.1644598421.dante.2344752.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e08300d3291d9aad128e41174c495ea05018dd7ad3834d0dcfbb5303cb1fda39
+size 5008

runs/Feb11_17-40-48_dante/events.out.tfevents.1644599097.dante.2344752.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1729f61fe3364a93e9a9d1f66da5ee6cdf7d7f562da2b58868f2f427d68e93f
+size 364

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "./", "tokenizer_class": "Wav2Vec2CTCTokenizer", "processor_class": "Wav2Vec2ProcessorWithLM"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74c7011121d1be25821c63534830c66a1646d322c4dc925d27818c2ade5a1e67
+size 2991

vocab.json

@@ -0,0 +1 @@
+{"a": 1, "b": 2, "c": 3, "d": 4, "e": 5, "f": 6, "g": 7, "h": 8, "i": 9, "j": 10, "k": 11, "l": 12, "m": 13, "n": 14, "o": 15, "p": 16, "q": 17, "r": 18, "s": 19, "t": 20, "u": 21, "v": 22, "w": 23, "x": 24, "y": 25, "z": 26, "å": 27, "æ": 28, "ø": 29, "|": 0, "[UNK]": 30, "[PAD]": 31}