Whisper Large V3 Turbo Turkish

🚀 Whisper Large v3 Turbo TR - Selim Çavaş

This model is a fine - tuned version of openai/whisper-large-v3-turbo on the Common Voice 17.0 dataset. It can be used for Turkish speech recognition tasks, achieving high accuracy in transcribing Turkish audio.

🚀 Quick Start

Basic Usage

import torch
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline

device = "cuda:0" if torch.cuda.is_available() else "cpu"
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32

model_id = "selimc/whisper-large-v3-turbo-turkish"

model = AutoModelForSpeechSeq2Seq.from_pretrained(
    model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True
)
model.to(device)

processor = AutoProcessor.from_pretrained(model_id)

pipe = pipeline(
    "automatic-speech-recognition",
    model=model,
    tokenizer=processor.tokenizer,
    feature_extractor=processor.feature_extractor,
    chunk_length_s=30,
    batch_size=16,
    return_timestamps=True,
    torch_dtype=torch_dtype,
    device=device,
)

result = pipe("test.mp3")
print(result["text"])

✨ Features

• Versatile Applications: This model can be used in various application areas, including transcription of Turkish language, voice commands, and automatic subtitling for Turkish videos.
• High Accuracy: Achieves a Wer of 18.9229 on the test set of the Common Voice 17.0 dataset.

📦 Installation

The code example above assumes you have installed the necessary libraries. You can install them using the following command:

pip install transformers torch datasets tokenizers

📚 Documentation

Intended uses & limitations

This model can be used in various application areas, including:

• Transcription of Turkish language
• Voice commands
• Automatic subtitling for Turkish videos

How To Use

The code example in the "Quick Start" section demonstrates how to use the model for automatic speech recognition.

Training

Due to colab GPU constraints, only 25% of the Turkish data available in the Common Voice 17.0 dataset was used for training. 😔

Got a GPU to spare? Let's collaborate and take this model to the next level! 🚀

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e - 05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e - 08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.1

📄 License

This model is released under the MIT license.

📦 Model Information