patrickvonplaten commited on
Commit
5e3d456
1 Parent(s): 7717449

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -8
README.md CHANGED
@@ -44,19 +44,12 @@ To transcribe audio files the model can be used as a standalone acoustic model a
44
  # load model and tokenizer
45
  processor = Wav2Vec2Processor.from_pretrained("facebook/wav2vec2-base-960h")
46
  model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-base-960h")
47
-
48
- # define function to read in sound file
49
- def map_to_array(batch):
50
- speech, _ = sf.read(batch["file"])
51
- batch["speech"] = speech
52
- return batch
53
 
54
  # load dummy dataset and read soundfiles
55
  ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
56
- ds = ds.map(map_to_array)
57
 
58
  # tokenize
59
- input_values = processor(ds["speech"][:2], return_tensors="pt", padding="longest").input_values # Batch size 1
60
 
61
  # retrieve logits
62
  logits = model(input_values).logits
 
44
  # load model and tokenizer
45
  processor = Wav2Vec2Processor.from_pretrained("facebook/wav2vec2-base-960h")
46
  model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-base-960h")
 
 
 
 
 
 
47
 
48
  # load dummy dataset and read soundfiles
49
  ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
 
50
 
51
  # tokenize
52
+ input_values = processor(ds[0]["audio"]["array"], return_tensors="pt", padding="longest").input_values # Batch size 1
53
 
54
  # retrieve logits
55
  logits = model(input_values).logits