add model

README.md

@@ -0,0 +1,54 @@
+---
+language:
+- ru
+license: apache-2.0
+tags:
+- automatic-speech-recognition
+- mozilla-foundation/common_voice_8_0
+- ru
+- robust-speech-event
+datasets:
+- mozilla-foundation/common_voice_8_0
+model-index:
+- name: XLS-R-1B - Russian
+  results:
+  - task: 
+      name: Automatic Speech Recognition 
+      type: automatic-speech-recognition
+    dataset:
+      name: Common Voice 8
+      type: mozilla-foundation/common_voice_8_0
+      args: ru
+    metrics:
+       - name: Test WER
+         type: wer
+         value: 9.87
+       - name: Test CER
+         type: cer
+         value: 2.32
+       - name: Test WER (+LM)
+         type: wer
+         value: 
+       - name: Test CER (+LM)
+         type: cer
+         value: 
+---
+
+# XLS-R-1B-RUSSIAN
+
+This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - RU dataset.
+
+
+## Evaluation Commands
+
+1. To evaluate on `mozilla-foundation/common_voice_8_0` with split `test`
+
+```bash
+python eval.py --model_id jonatasgrosman/wav2vec2-xls-r-1b-russian --dataset mozilla-foundation/common_voice_8_0 --config ru --split test
+```
+
+2. To evaluate on `speech-recognition-community-v2/dev_data`
+
+```bash
+python eval.py --model_id jonatasgrosman/wav2vec2-xls-r-1b-russian --dataset speech-recognition-community-v2/dev_data --config ru --split validation --chunk_length_s 5.0 --stride_length_s 1.0
+```

alphabet.json

@@ -0,0 +1 @@
+{"labels": ["", "<s>", "</s>", "\u2047", " ", "'", "-", "\u0430", "\u0431", "\u0432", "\u0433", "\u0434", "\u0435", "\u0436", "\u0437", "\u0438", "\u0439", "\u043a", "\u043b", "\u043c", "\u043d", "\u043e", "\u043f", "\u0440", "\u0441", "\u0442", "\u0443", "\u0444", "\u0445", "\u0446", "\u0447", "\u0448", "\u0449", "\u044a", "\u044b", "\u044c", "\u044d", "\u044e", "\u044f", "\u0451"], "is_bpe": false}

config.json

@@ -0,0 +1,107 @@
+{
+  "_name_or_path": "facebook/wav2vec2-xls-r-1b",
+  "activation_dropout": 0.05,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.05,
+  "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": false,
+  "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": 0.05,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.05,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.05,
+  "hidden_size": 1280,
+  "initializer_range": 0.02,
+  "intermediate_size": 5120,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.05,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.05,
+  "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": 0,
+  "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": "4.16.0.dev0",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 40,
+  "xvector_output_dim": 512
+}

eval.py

@@ -0,0 +1,146 @@
+#!/usr/bin/env python3
+from datasets import load_dataset, load_metric, Audio, Dataset
+from transformers import pipeline, AutoFeatureExtractor, AutoTokenizer
+import re
+import torch
+import argparse
+from typing import Dict
+
+
+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, invalid_chars_regex: str, to_lower: bool) -> str:
+    """ DO ADAPT FOR YOUR USE CASE. this function normalizes the target text. """
+
+    text = text.lower() if to_lower else text.upper()
+
+    text = re.sub(invalid_chars_regex, " ", text)
+
+    text = re.sub("\s+", " ", text).strip()
+
+    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)
+
+    # build normalizer config
+    tokenizer = AutoTokenizer.from_pretrained(args.model_id)
+    tokens = [x for x in tokenizer.convert_ids_to_tokens(range(0, tokenizer.vocab_size))]
+    special_tokens = [
+        tokenizer.pad_token, tokenizer.word_delimiter_token,
+        tokenizer.unk_token, tokenizer.bos_token,
+        tokenizer.eos_token,
+    ]
+    non_special_tokens = [x for x in tokens if x not in special_tokens]
+    invalid_chars_regex = f"[^\s{re.escape(''.join(set(non_special_tokens)))}]"
+    normalize_to_lower = False
+    for token in non_special_tokens:
+        if token.isalpha() and token.islower():
+            normalize_to_lower = True
+            break
+
+    # map function to decode audio
+    def map_to_pred(batch, args=args, asr=asr, invalid_chars_regex=invalid_chars_regex, normalize_to_lower=normalize_to_lower):
+        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"], invalid_chars_regex, normalize_to_lower)
+        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 None. For long audio files a good value would be 5.0 seconds."
+    )
+    parser.add_argument(
+        "--stride_length_s", type=float, default=None, help="Stride of the audio chunks. Defaults to None. For long audio files a good value would be 1.0 seconds."
+    )
+    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/attrs.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5ffd02e1ceef6517476e72ebe7997ddef7e92d27cb5a23d6695d64c4317d6ad
+size 78

language_model/lm.binary

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01a8fba99ad0d5157e060ea7bee83adb7bc7025606e6f98a783da1413b6edbaf
+size 1486935543

language_model/unigrams.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7c72ae916afa54951fe3bd6094c95e2895c1c7881a20e02ef640aa27cdb4874
+size 32849086

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:d0e528a0026d705ac327145b582edb504aacde9962c67a602525c7c12cdc3c77
+size 3850517681

special_tokens_map.json

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

vocab.json

@@ -0,0 +1 @@
+{"<pad>": 0, "<s>": 1, "</s>": 2, "<unk>": 3, "|": 4, "'": 5, "-": 6, "а": 7, "б": 8, "в": 9, "г": 10, "д": 11, "е": 12, "ж": 13, "з": 14, "и": 15, "й": 16, "к": 17, "л": 18, "м": 19, "н": 20, "о": 21, "п": 22, "р": 23, "с": 24, "т": 25, "у": 26, "ф": 27, "х": 28, "ц": 29, "ч": 30, "ш": 31, "щ": 32, "ъ": 33, "ы": 34, "ь": 35, "э": 36, "ю": 37, "я": 38, "ё": 39}