Ru Whisper Small

🚀 ru_whisper_small - Val123val

This model is a fine - tuned version of openai/whisper-small on the Sberdevices_golos_10h_crowd dataset, which is potentially quite useful as an ASR solution, especially for Russian speech recognition.

✨ Features

• Transformer - based: Whisper is a Transformer based encoder - decoder (sequence - to - sequence) model, trained on 680k hours of labelled speech data with large - scale weak supervision. The Russian language data accounts for only 5k hours of the total.
• Fine - tuned: ru_whisper_small is a fine - tuned version of openai/whisper-small on the Sberdevices_golos_10h_crowd dataset, offering enhanced performance for specific tasks.
• Long - Form Transcription: Can transcribe audio of arbitrary length using a chunking algorithm.
• Speculative Decoding: Supports speculative decoding for faster inference.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

from transformers import WhisperProcessor, WhisperForConditionalGeneration
from datasets import load_dataset

# load model and processor
processor = WhisperProcessor.from_pretrained("Val123val/ru_whisper_small")
model = WhisperForConditionalGeneration.from_pretrained("Val123val/ru_whisper_small")
model.config.forced_decoder_ids = None

# load dataset and read audio files
ds = load_dataset("bond005/sberdevices_golos_10h_crowd", split="validation", token=True)
sample = ds[0]["audio"]
input_features = processor(sample["array"], sampling_rate=sample["sampling_rate"], return_tensors="pt").input_features 

# generate token ids
predicted_ids = model.generate(input_features)
# decode token ids to text
transcription = processor.batch_decode(predicted_ids, skip_special_tokens=False)

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

Advanced Usage - Long - Form Transcription

import torch
from transformers import pipeline
from datasets import load_dataset

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

pipe = pipeline(
  "automatic-speech-recognition",
  model="Val123val/ru_whisper_small",
  chunk_length_s=30,
  device=device,
)

ds = load_dataset("bond005/sberdevices_golos_10h_crowd", split="validation", token=True)
sample = ds[0]["audio"]

prediction = pipe(sample.copy(), batch_size=8)["text"]

# we can also return timestamps for the predictions
prediction = pipe(sample.copy(), batch_size=8, return_timestamps=True)["chunks"]

Advanced Usage - Faster using with Speculative Decoding

import torch
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor
from transformers import pipeline

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

# load dataset
dataset = load_dataset("bond005/sberdevices_golos_10h_crowd", split="validation", token=True)

# load model
model_id = "Val123val/ru_whisper_small"

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

processor = AutoProcessor.from_pretrained(model_id)

# load assistant model
assistant_model_id = "openai/whisper-tiny"

assistant_model = AutoModelForSpeechSeq2Seq.from_pretrained(
    assistant_model_id,
    torch_dtype=torch_dtype,
    low_cpu_mem_usage=True,
    use_safetensors=True,
    attn_implementation="sdpa",
)

assistant_model.to(device);

# make pipe
pipe = pipeline(
    "automatic-speech-recognition",
    model=model,
    tokenizer=processor.tokenizer,
    feature_extractor=processor.feature_extractor,
    max_new_tokens=128,
    chunk_length_s=15,
    batch_size=4,
    generate_kwargs={"assistant_model": assistant_model},
    torch_dtype=torch_dtype,
    device=device,
)

sample = dataset[0]["audio"]
result = pipe(sample)
print(result["text"])

📚 Documentation

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 32
• 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

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.0
• Tokenizers 0.15.0

📄 License

This project is licensed under the Apache - 2.0 license.