Upload lm-boosted decoder

.gitattributes

@@ -25,3 +25,4 @@ 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
+language_model/unigrams.txt filter=lfs diff=lfs merge=lfs -text

.ipynb_checkpoints/run-checkpoint.sh

@@ -1,33 +0,0 @@
-python run_speech_recognition_ctc.py \
-        --dataset_name="mozilla-foundation/common_voice_8_0" \
-        --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
-        --dataset_config_name="mr" \
-        --output_dir="./" \
-        --overwrite_output_dir \
-        --num_train_epochs="500" \
-        --per_device_train_batch_size="32" \
-        --per_device_eval_batch_size="16" \
-        --learning_rate="7.5e-5" \
-        --warmup_steps="2000" \
-        --length_column_name="input_length" \
-        --evaluation_strategy="steps" \
-        --text_column_name="sentence" \
-        --save_steps="400" \
-        --eval_steps="400" \
-        --logging_steps="100" \
-        --layerdrop="0.0" \
-        --activation_dropout="0.1" \
-        --save_total_limit="1" \
-        --freeze_feature_encoder \
-        --feat_proj_dropout="0.0" \
-        --mask_time_prob="0.75" \
-        --mask_time_length="10" \
-        --mask_feature_prob="0.25" \
-        --mask_feature_length="64" \
-        --chars_to_ignore , ? . ! \- \; \: \" “ % ‘ ” � — ’ … – \' । \॔ \
-        --gradient_checkpointing \
-        --use_auth_token \
-        --fp16 \
-        --group_by_length \
-        --do_train --do_eval \
-        --push_to_hub

.ipynb_checkpoints/vocab-checkpoint.json

@@ -1 +0,0 @@
-{"ँ": 1, "ं": 2, "ः": 3, "अ": 4, "आ": 5, "इ": 6, "ई": 7, "उ": 8, "ऊ": 9, "ऋ": 10, "ए": 11, "ऐ": 12, "ऑ": 13, "ओ": 14, "औ": 15, "क": 16, "ख": 17, "ग": 18, "घ": 19, "च": 20, "छ": 21, "ज": 22, "झ": 23, "ञ": 24, "ट": 25, "ठ": 26, "ड": 27, "ढ": 28, "ण": 29, "त": 30, "थ": 31, "द": 32, "ध": 33, "न": 34, "प": 35, "फ": 36, "ब": 37, "भ": 38, "म": 39, "य": 40, "र": 41, "ऱ": 42, "ल": 43, "ळ": 44, "व": 45, "श": 46, "ष": 47, "स": 48, "ह": 49, "़": 50, "ा": 51, "ि": 52, "ी": 53, "ु": 54, "ू": 55, "ृ": 56, "ॄ": 57, "ॅ": 58, "े": 59, "ै": 60, "ॉ": 61, "ॊ": 62, "ो": 63, "ौ": 64, "्": 65, "|": 0, "[UNK]": 66, "[PAD]": 67}

alphabet.json

@@ -0,0 +1 @@
+{"labels": [" ", "\u0901", "\u0902", "\u0903", "\u0905", "\u0906", "\u0907", "\u0908", "\u0909", "\u090a", "\u090b", "\u090f", "\u0910", "\u0911", "\u0913", "\u0914", "\u0915", "\u0916", "\u0917", "\u0918", "\u091a", "\u091b", "\u091c", "\u091d", "\u091e", "\u091f", "\u0920", "\u0921", "\u0922", "\u0923", "\u0924", "\u0925", "\u0926", "\u0927", "\u0928", "\u092a", "\u092b", "\u092c", "\u092d", "\u092e", "\u092f", "\u0930", "\u0931", "\u0932", "\u0933", "\u0935", "\u0936", "\u0937", "\u0938", "\u0939", "\u093c", "\u093e", "\u093f", "\u0940", "\u0941", "\u0942", "\u0943", "\u0944", "\u0945", "\u0947", "\u0948", "\u0949", "\u094a", "\u094b", "\u094c", "\u094d", "\u2047", "", "<s>", "</s>"], "is_bpe": false}

eval.py

@@ -0,0 +1,164 @@
+#!/usr/bin/env python3
+import argparse
+import re
+import unicodedata
+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 = unicodedata.normalize("NFKC", text)
+    text = re.sub(chars_to_ignore_regex, "", text.lower())
+
+    # 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["sentence"])
+        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)

language_model/4gram.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4638564096b5800133f9de231905227633addc987b5a5e5ee7c2e3e64892e802
+size 5301634441

language_model/attrs.json

@@ -0,0 +1 @@
+{"alpha": 0.5, "beta": 1.5, "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 sha256:475211592ab58f3029e5120c4fb87960d5e909472c2e2fc948393275a4b3d44a
+size 116468766

preprocessor_config.json

@@ -4,6 +4,7 @@
   "feature_size": 1,
   "padding_side": "right",
   "padding_value": 0,
+  "processor_class": "Wav2Vec2ProcessorWithLM",
   "return_attention_mask": true,
   "sampling_rate": 16000
 }

special_tokens_map.json

@@ -1 +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}]}
+{"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}]}

tokenizer_config.json

@@ -1 +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, "name_or_path": "./", "tokenizer_class": "Wav2Vec2CTCTokenizer"}
+{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": null, "name_or_path": "anuragshas/wav2vec2-xls-r-300m-mr-cv8-with-lm", "tokenizer_class": "Wav2Vec2CTCTokenizer", "processor_class": "Wav2Vec2ProcessorWithLM"}