Whisper Large V2 Japanese 5k Steps

🚀 whisper-large-v2-japanese-5k-steps

This model is a fine - tuned version of openai/whisper-large-v2 on the Japanese CommonVoice dataset (v11). It aims to transcribe Japanese speech, and although it's for research, it provides certain performance metrics on the evaluation set.

🚀 Quick Start

This model is a fine - tuned version of openai/whisper-large-v2 on the Japanese CommonVoice dataset (v11). It achieves the following results on the evaluation set:

• Loss: 0.4200
• Wer: 0.7449

✨ Features

This model is finetuned for 5000 steps for research purposes which means that the transcriptions might not be that satisfactory for users.

📦 Installation

No specific installation steps are provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

from datasets import load_dataset, Audio
import torch
from transformers import WhisperProcessor, WhisperForConditionalGeneration

# device
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

# load the model
processor = WhisperProcessor.from_pretrained("clu-ling/whisper-large-v2-japanese-5k-steps")
model = WhisperForConditionalGeneration.from_pretrained("clu-ling/whisper-large-v2-japanese-5k-steps").to(device)
forced_decoder_ids = processor.get_decoder_prompt_ids(language="ja", task="transcribe")

# load the dataset
commonvoice_eval = load_dataset("mozilla-foundation/common_voice_11_0", "ja", split="validation", streaming=True)
commonvoice_eval = commonvoice_eval.cast_column("audio", Audio(sampling_rate=16000))
sample = next(iter(commonvoice_eval))["audio"]

# features and generate token ids
input_features = processor(sample["array"], sampling_rate=sample["sampling_rate"], return_tensors="pt").input_features
predicted_ids = model.generate(input_features.to(device), forced_decoder_ids=forced_decoder_ids)

# decode
transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)

print(transcription)

Advanced Usage

from transformers.models.whisper.english_normalizer import BasicTextNormalizer
from datasets import load_dataset, Audio
import evaluate
import torch
import re
from transformers import WhisperProcessor, WhisperForConditionalGeneration

# device
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

# metric
wer_metric = evaluate.load("wer")

# model
processor = WhisperProcessor.from_pretrained("clu-ling/whisper-large-v2-japanese-5k-steps")
model = WhisperForConditionalGeneration.from_pretrained("clu-ling/whisper-large-v2-japanese-5k-steps")

# dataset
dataset = load_dataset("mozilla-foundation/common_voice_11_0", "ja", split="test", ) #cache_dir=args.cache_dir
dataset = dataset.cast_column("audio", Audio(sampling_rate=16000))

#for debuggings: it gets some examples
#dataset = dataset.shard(num_shards=7000, index=0)
#print(dataset)
   
def normalize(batch):
  batch["gold_text"] = whisper_norm(batch['sentence'])
  return batch

def map_wer(batch):
  model.to(device)
  forced_decoder_ids = processor.get_decoder_prompt_ids(language = "ja", task = "transcribe")
  inputs = processor(batch["audio"]["array"], sampling_rate=batch["audio"]["sampling_rate"], return_tensors="pt").input_features
  with torch.no_grad():
    generated_ids = model.generate(inputs=inputs.to(device), forced_decoder_ids=forced_decoder_ids)
    transcription = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
  batch["predicted_text"] = whisper_norm(transcription)
  return batch

# process GOLD text
processed_dataset = dataset.map(normalize)
# get predictions
predicted = processed_dataset.map(map_wer)

# word error rate
wer = wer_metric.compute(references=predicted['gold_text'], predictions=predicted['predicted_text'])
wer = round(100 * wer, 2)
print("WER:", wer)

📚 Documentation

Training and evaluation data

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e - 05
• train_batch_size: 50
• eval_batch_size: 16
• 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: 5000
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.1
• Datasets 2.8.1.dev0
• Tokenizers 0.13.2

🔧 Technical Details

The model is based on fine - tuning openai/whisper-large-v2 with 5000 steps of training on the Japanese CommonVoice dataset (v11). The hyperparameters and training process are described above, which are used to optimize the model's performance on Japanese speech transcription.

📄 License

This model is released under the Apache - 2.0 license.