librarian-bot's picture
Librarian Bot: Add base_model information to model
c3db3b3
metadata
language: id
license: apache-2.0
tags:
  - audio
  - automatic-speech-recognition
  - speech
  - xlsr-fine-tuning-week
datasets:
  - common_voice
metrics:
  - wer
base_model: facebook/wav2vec2-large-xlsr-53
model-index:
  - name: XLSR Wav2Vec2 Indonesian Mix by Cahya
    results:
      - task:
          type: automatic-speech-recognition
          name: Speech Recognition
        dataset:
          name: Common Voice id
          type: common_voice
          args: id
        metrics:
          - type: wer
            value: 19.36
            name: Test WER

Wav2Vec2-Large-XLSR-Indonesian

Fine-tuned facebook/wav2vec2-large-xlsr-53 on the Indonesian Common Voice dataset and synthetic voices generated using Artificial Common Voicer, which again based on Google Text To Speech. When using this model, make sure that your speech input is sampled at 16kHz.

Usage

The model can be used directly (without a language model) as follows:

import torch
import torchaudio
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor

test_dataset = load_dataset("common_voice", "id", split="test[:2%]")

processor = Wav2Vec2Processor.from_pretrained("cahya/wav2vec2-large-xlsr-indonesian-mix")
model = Wav2Vec2ForCTC.from_pretrained("cahya/wav2vec2-large-xlsr-indonesian-mix")


# Preprocessing the datasets.
# We need to read the aduio files as arrays
def speech_file_to_array_fn(batch):
    speech_array, sampling_rate = torchaudio.load(batch["path"])
    resampler = torchaudio.transforms.Resample(sampling_rate, 16_000)
    batch["speech"] = resampler(speech_array).squeeze().numpy()
    return batch

test_dataset = test_dataset.map(speech_file_to_array_fn)
inputs = processor(test_dataset[:2]["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)

with torch.no_grad():
    logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits

predicted_ids = torch.argmax(logits, dim=-1)

print("Prediction:", processor.batch_decode(predicted_ids))
print("Reference:", test_dataset[:2]["sentence"])

Evaluation

The model can be evaluated as follows on the Indonesian test data of Common Voice.

import torch
import torchaudio
from datasets import load_dataset, load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
import re

test_dataset = load_dataset("common_voice", "id", split="test")
wer = load_metric("wer")

processor = Wav2Vec2Processor.from_pretrained("cahya/wav2vec2-large-xlsr-indonesian-mix")
model = Wav2Vec2ForCTC.from_pretrained("cahya/wav2vec2-large-xlsr-indonesian-mix") 
model.to("cuda")

chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“\%\‘\'\”\�]'


# Preprocessing the datasets.
# We need to read the aduio files as arrays
def speech_file_to_array_fn(batch):
    batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
    speech_array, sampling_rate = torchaudio.load(batch["path"])
    resampler = torchaudio.transforms.Resample(sampling_rate, 16_000)
    batch["speech"] = resampler(speech_array).squeeze().numpy()
    return batch

test_dataset = test_dataset.map(speech_file_to_array_fn)

# Preprocessing the datasets.
# We need to read the aduio files as arrays
def evaluate(batch):
    inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)

    with torch.no_grad():
        logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits

    pred_ids = torch.argmax(logits, dim=-1)
    batch["pred_strings"] = processor.batch_decode(pred_ids)
    return batch

result = test_dataset.map(evaluate, batched=True, batch_size=8)

print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))

Test Result: 19.36 %

Training

The Common Voice train, validation, and ... datasets were used for training as well as ... and ... # TODO

The script used for training can be found here