Whisper Base.kk

🚀 Whisper

Whisper-base for automatic speech recognition (ASR) tailored for the low - resourced Kazakh language. This model offers an effective solution for transcribing Kazakh speech, leveraging extensive training data.

🚀 Quick Start

This model is a fine - tuned Whisper - base for automatic speech recognition of the Kazakh language. It was trained on the Kazakh Speech Corpus 2 with more than 1k hours of labeled data and achieved a 15.36% WER on the test set.

✨ Features

• Language - Specific: It is a Kazakh - only model, dedicated to high - quality Kazakh speech recognition.
• Long - Form Transcription: Through a chunking algorithm, it can transcribe audio of arbitrary length.

📦 Installation

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

💻 Usage Examples

Basic Usage

>>> from transformers import WhisperProcessor, WhisperForConditionalGeneration
>>> import librosa

>>> # load model and processor
>>> processor = WhisperProcessor.from_pretrained("akuzdeuov/whisper-base.kk")
>>> model = WhisperForConditionalGeneration.from_pretrained("akuzdeuov/whisper-base.kk")

>>> # load your audio
>>> audio, sampling_rate = librosa.load("path_to_audio", sr=16000)
>>> input_features = processor(audio, sampling_rate=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)

The context tokens can be removed from the start of the transcription by setting skip_special_tokens=True.

Advanced Usage

>>> import torch
>>> from transformers import pipeline

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

>>> pipe = pipeline(
>>>   "automatic-speech-recognition",
>>>   model="akuzdeuov/whisper-base.kk",
>>>   chunk_length_s=30,
>>>   device=device,
>>> )

>>> prediction = pipe("path_to_audio", batch_size=8)["text"]

The Whisper model is intrinsically designed to work on audio samples of up to 30s in duration. However, by using a chunking algorithm, it can be used to transcribe audio samples of up to arbitrary length. This is possible through Transformers pipeline method. Chunking is enabled by setting chunk_length_s = 30 when instantiating the pipeline. With chunking enabled, the pipeline can be run with batched inference.

📚 Documentation

🔧 Technical Details

No detailed technical implementation details are provided in the original document, so this section is skipped.

📄 License

This project is licensed under the apache - 2.0 license.

📚 References

1. Whisper, OpenAI.