Wav2vec2 Large Xlsr Turkish Artificial Cv

🚀 Wav2Vec2-Large-XLSR-Turkish

This is a fine - tuned model for speech recognition, specifically designed for the Turkish language. It is based on the Wav2Vec2 - Large - XLSR architecture and trained on the Turkish Common Voice dataset.

🚀 Quick Start

This is the model for Wav2Vec2-Large-XLSR-Turkish-Artificial-CV, a fine-tuned cahya/wav2vec2-large-xlsr-turkish-artificial model on Turkish Common Voice dataset.

When using this model, make sure that your speech input is sampled at 16kHz.

✨ Features

• Language: Turkish
• Datasets: Common Voice
• Metrics: Word Error Rate (WER)
• Tags: audio, automatic - speech - recognition, speech, xlsr - fine - tuning - week
• License: Apache 2.0

Model Index:

• Name: XLSR Wav2Vec2 Turkish by Cahya Results:
  • Task: Name: Speech Recognition Type: automatic - speech - recognition Dataset: Name: Common Voice tr Type: common_voice Args: tr Metrics:
    • Name: Test WER Type: wer Value: 14.61

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

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

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

processor = Wav2Vec2Processor.from_pretrained("cahya/wav2vec2-large-xlsr-turkish-artificial-cv")
model = Wav2Vec2ForCTC.from_pretrained("cahya/wav2vec2-large-xlsr-turkish-artificial-cv")


# 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"])

Advanced Usage

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

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

processor = Wav2Vec2Processor.from_pretrained("cahya/wav2vec2-large-xlsr-turkish-artificial-cv")
model = Wav2Vec2ForCTC.from_pretrained("cahya/wav2vec2-large-xlsr-turkish-artificial-cv") 
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: 14.61 %

📚 Documentation

The Common Voice train, validation, other and invalidated

The script used for training can be found here

📄 License

This model is licensed under the Apache 2.0 license.