.gitattributes

@@ -14,4 +14,3 @@
 *.pb filter=lfs diff=lfs merge=lfs -text
 *.pt filter=lfs diff=lfs merge=lfs -text
 *.pth filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text

README.md

@@ -6,39 +6,7 @@ tags:
 - speech
 - audio
 - automatic-speech-recognition
-- hf-asr-leaderboard
 license: apache-2.0
-model-index:
-- name: wav2vec2-large-960h-lv60
-  results:
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: LibriSpeech (clean)
-      type: librispeech_asr
-      config: clean
-      split: test
-      args: 
-        language: en
-    metrics:
-    - name: Test WER
-      type: wer
-      value: 1.9
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: LibriSpeech (other)
-      type: librispeech_asr
-      config: other
-      split: test
-      args: 
-        language: en
-    metrics:
-    - name: Test WER
-      type: wer
-      value: 3.9
 ---
 
 # Wav2Vec2-Large-960h-Lv60 + Self-Training
@@ -63,26 +31,34 @@ The original model can be found under https://github.com/pytorch/fairseq/tree/ma
 To transcribe audio files the model can be used as a standalone acoustic model as follows:
 
 ```python
- from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC
+ from transformers import Wav2Vec2Tokenizer, Wav2Vec2ForCTC
  from datasets import load_dataset
+ import soundfile as sf
  import torch
  
- # load model and processor
+ # load model and tokenizer
- processor = Wav2Vec2Processor.from_pretrained("facebook/wav2vec2-large-960h-lv60-self")
+ tokenizer = Wav2Vec2Tokenizer.from_pretrained("facebook/wav2vec2-large-960h-lv60-self")
  model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-large-960h-lv60-self")
+ 
+ # define function to read in sound file
+ def map_to_array(batch):
+     speech, _ = sf.read(batch["file"])
+     batch["speech"] = speech
+     return batch
      
  # load dummy dataset and read soundfiles
  ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
+ ds = ds.map(map_to_array)
  
  # tokenize
- input_values = processor(ds[0]["audio"]["array"], return_tensors="pt", padding="longest").input_values
+ input_values = tokenizer(ds["speech"][:2], return_tensors="pt", padding="longest").input_values  # Batch size 1
  
  # retrieve logits
  logits = model(input_values).logits
  
  # take argmax and decode
  predicted_ids = torch.argmax(logits, dim=-1)
- transcription = processor.batch_decode(predicted_ids)
+ transcription = tokenizer.batch_decode(predicted_ids)
  ```
  
   ## Evaluation
@@ -91,7 +67,8 @@ To transcribe audio files the model can be used as a standalone acoustic model a
  
 ```python
 from datasets import load_dataset
-from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
+from transformers import Wav2Vec2ForCTC, Wav2Vec2Tokenizer
+import soundfile as sf
 import torch
 from jiwer import wer
 
@@ -99,10 +76,17 @@ from jiwer import wer
 librispeech_eval = load_dataset("librispeech_asr", "clean", split="test")
 
 model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-large-960h-lv60-self").to("cuda")
-processor = Wav2Vec2Processor.from_pretrained("facebook/wav2vec2-large-960h-lv60-self")
+tokenizer = Wav2Vec2Tokenizer.from_pretrained("facebook/wav2vec2-large-960h-lv60-self")
+
+def map_to_array(batch):
+    speech, _ = sf.read(batch["file"])
+    batch["speech"] = speech
+    return batch
+
+librispeech_eval = librispeech_eval.map(map_to_array)
 
 def map_to_pred(batch):
-    inputs = processor(batch["audio"]["array"], return_tensors="pt", padding="longest")
+    inputs = tokenizer(batch["speech"], return_tensors="pt", padding="longest")
     input_values = inputs.input_values.to("cuda")
     attention_mask = inputs.attention_mask.to("cuda")
     
@@ -110,11 +94,11 @@ def map_to_pred(batch):
         logits = model(input_values, attention_mask=attention_mask).logits
 
     predicted_ids = torch.argmax(logits, dim=-1)
-    transcription = processor.batch_decode(predicted_ids)
+    transcription = tokenizer.batch_decode(predicted_ids)
     batch["transcription"] = transcription
     return batch
 
-result = librispeech_eval.map(map_to_pred, remove_columns=["audio"])
+result = librispeech_eval.map(map_to_pred, batched=True, batch_size=16, remove_columns=["speech"])
 
 print("WER:", wer(result["text"], result["transcription"]))
 ```

config.json

@@ -1,14 +1,7 @@
 {
-  "_name_or_path": "facebook/wav2vec2-large-960h-lv60-self",
-  "activation_dropout": 0.1,
-  "apply_spec_augment": true,
   "architectures": [
     "Wav2Vec2ForCTC"
   ],
-  "attention_dropout": 0.1,
-  "bos_token_id": 1,
-  "codevector_dim": 256,
-  "contrastive_logits_temperature": 0.1,
   "conv_bias": true,
   "conv_dim": [
     512,
@@ -37,41 +30,22 @@
     2,
     2
   ],
-  "ctc_loss_reduction": "sum",
-  "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": 0.1,
-  "feat_quantizer_dropout": 0.0,
-  "final_dropout": 0.1,
-  "gradient_checkpointing": false,
   "hidden_act": "gelu",
-  "hidden_dropout": 0.1,
   "hidden_dropout_prob": 0.1,
   "hidden_size": 1024,
   "initializer_range": 0.02,
   "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
-  "layerdrop": 0.1,
-  "mask_feature_length": 10,
-  "mask_feature_prob": 0.0,
-  "mask_time_length": 10,
-  "mask_time_prob": 0.05,
   "model_type": "wav2vec2",
   "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": 24,
-  "num_negatives": 100,
+  "transformers_version": "4.3.0.dev0",
-  "pad_token_id": 0,
-  "proj_codevector_dim": 256,
-  "transformers_version": "4.7.0.dev0",
   "vocab_size": 32
 }