Update README.md

README.md

@@ -1,96 +1,61 @@
-# Wav2Vec2 Acoustic Model fine-tuned on LibriSpeech
+---
+language: en
+datasets:
+- librispeech_asr
+tags:
+- speech
 
-Original model can be found under https://github.com/pytorch/fairseq/tree/master/examples/wav2vec#wav2vec-20.
+license: apache-2.0
+---
 
-Paper: https://arxiv.org/abs/2006.11477
+# Wav2Vec2-Base-960h
 
-## Usage
+[Facebook's Wav2Vec2](https://ai.facebook.com/blog/wav2vec-20-learning-the-structure-of-speech-from-raw-audio/)
 
-Make sure you are working on [this branch](https://github.com/huggingface/transformers/tree/add_wav2vec) (which will be merged to master soon hopefully) of transformers:
+The large model pretrained and fine-tuned on 960 hours of Librispeech.
 
-```bash
-$ git checkout add_wav2vec
-```
+[Paper](https://arxiv.org/abs/2006.11477)
 
-In the following, we'll show a simple example of how the model can be used for automatic speech recognition.
+Authors: Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli
 
-First, let's load the model
+**Abstract**
 
-```python
-from transformers import AutoModelForMaskedLM
-
-model = AutoModelForMaskedLM.from_pretrained("patrickvonplaten/wav2vec2-large-960h")
-
-```
-
-Next, let's load a dummy librispeech dataset
-
-```python
-from datasets import load_dataset
-import soundfile as sf
-
-libri_speech_dummy = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
-
-def map_to_array(batch):
-    speech_array, _ = sf.read(batch["file"])
-    batch["speech"] = speech_array
-    return batch
-
-libri_speech_dummy = libri_speech_dummy.map(map_to_array, remove_columns=["file"])
-
-# check out dataset
-print(libri_speech_dummy)
+We show for the first time that learning powerful representations from speech audio alone followed by fine-tuning on transcribed speech can outperform the best semi-supervised methods while being conceptually simpler. wav2vec 2.0 masks the speech input in the latent space and solves a contrastive task defined over a quantization of the latent representations which are jointly learned. Experiments using all labeled data of Librispeech achieve 1.8/3.3 WER on the clean/other test sets. When lowering the amount of labeled data to one hour, wav2vec 2.0 outperforms the previous state of the art on the 100 hour subset while using 100 times less labeled data. Using just ten minutes of labeled data and pre-training on 53k hours of unlabeled data still achieves 4.8/8.2 WER. This demonstrates the feasibility of speech recognition with limited amounts of labeled data.
 
-input_speech_16kHz = libri_speech_dummy[2]["speech"]
-expected_trans = libri_speech_dummy[2]["text"]
-```
+The original model can be found under https://github.com/pytorch/fairseq/tree/master/examples/wav2vec#wav2vec-20.
 
-Cool, now we can run an inference pass to retrieve the logits:
 
-```python
-import torch
-logits = model(torch.tensor(input_speech_16kHz)[None, :])
-
-# use highest probability logits
-pred_ids = torch.argmax(logits[0], axis=-1)
-```
-
-Finally, let's decode the prediction. 
-Let's create a simple CTC-Decoder:
-
-```python
-  import numpy as np
-  from itertools import groupby
+# Usage
 
-  class Decoder:
-      def __init__(self, json_dict):
-          self.dict = json_dict
-          self.look_up = np.asarray(list(self.dict.keys()))
-  
-      def decode(self, ids):
-          converted_tokens = self.look_up[ids]
-          fused_tokens = [tok[0] for tok in groupby(converted_tokens)]
-          output = ' '.join(''.join(''.join(fused_tokens).split("<s>")).split("|"))
-          return output
-```
-
-and instantiate with the corresponding dict.
+To transcribe audio files the model can be used as a standalone acoustic model as follows:
 
 ```python
-# hard-coded json dict taken from: https://dl.fbaipublicfiles.com/fairseq/wav2vec/dict.ltr.txt
-json_dict = {"<s>": 0, "<pad>": 1, "</s>": 2, "<unk>": 3, "|": 4, "E": 5, "T": 6, "A": 7, "O": 8, "N": 9, "I": 10, "H": 11, "S": 12, "R": 13, "D": 14, "L": 15, "U": 16, "M": 17, "W": 18, "C": 19, "F": 20, "G": 21, "Y": 22, "P": 23, "B": 24, "V": 25, "K": 26, "'": 27, "X": 28, "J": 29, "Q": 30, "Z": 31}
-
-decoder = Decoder(json_dict=json_dict)
-```
-
-and decode the result
-
-```python   
-pred_trans = decoder.decode(pred_ids)
-   
-print("Prediction:\n", pred_trans)
-print("\n" + 50 * "=" + "\n")
-print("Correct result:\n", expected_trans)
-```
-
-🎉
+ from transformers import Wav2Vec2Tokenizer, Wav2Vec2Model
+ from datasets import load_dataset
+ import soundfile as sf
+ import torch
+ 
+ # load model and tokenizer
+ tokenizer = Wav2Vec2Tokenizer.from_pretrained("facebook/wav2vec2-large-960h")
+ model = Wav2Vec2ForMaskedLM.from_pretrained("facebook/wav2vec2-large-960h")
+ 
+ # 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 = 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 = tokenizer.batch_decode(predicted_ids)
+ ```