Automatic-Speech-Recognition - GitLite

KO | EN

gitlite — search

#javascript #hacktoberfest #react #llm #go #machine-learning #rust #deep-learning #nodejs #java #kubernetes #vue #mcp #open-source

Automatic-Speech-Recognition

(★ 222)

🎧 Automatic Speech Recognition: DeepSpeech & Seq2Seq (TensorFlow)

파일 탐색기

최종 버전 다운로드 (.zip)

- - __init__.py
  - augmentation.py
  - spec_augment.py
- - __init__.py
  - batch_logger.py
  - distributed_model_checkpoint.py
- - __init__.py
  - audio.py
  - dataset.py
  - features.py
- - __init__.py
  - decoder.py
- - __init__.py
  - activations.py
  - distance.py
  - evaluate.py
- - __init__.py
  - feature_extractor.py
  - filter_banks.py
  - spectrogram.py
- - __init__.py
  - load.py
- - __init__.py
  - deepspeech.py
  - deepspeech2.py
- - __init__.py
  - ctc_pipeline.py
  - pipeline.py
- - __init__.py
  - alphabet-en.txt
  - alphabet-pl.txt
  - alphabet.py
- - __init__.py
  - consumer.py
  - utils.py
- __init__.py
- basic.py
- test.csv
- use_augmentation.py
- use_callbacks.py
- - test_spec_augment.py
- - test_batch_logger.py
  - test_distributed_model_checkpoint.py
- - test_audio.py
  - test_dataset.py
  - test_features.py
- - test_decoder.py
- - test_feature_extractor.py
  - test_filter_banks.py
  - test_spectrogram.py
- - test_load.py
- - test_deepspeech.py
  - test_deepspeech2.py
- - test_ctc_pipeline.py
- - test_alphabet.py
- - test_utils.py
- conftest.py
- sample-en.wav
- - test_transfer.py
  - transfer.py
  - utils.py
.gitignore
environment-gpu.yml
environment.yml
LICENSE.md
MANIFEST.in
README.md
setup.py

# 설치 가이드

1. 코드 내려받기

git clone https://github.com/rolczynski/Automatic-Speech-Recognition

깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.

cd Automatic-Speech-Recognition

방금 내려받은 프로젝트 폴더 안으로 이동합니다.

2. 공식 설치 스크립트

쉬움 추천

사전 준비물

Python 3 pip 명령어를 쓰려면 Python이 필요합니다.

pip install automatic-speech-recognition

PyPI에 배포된 패키지를 바로 설치합니다. 소스 클론이 필요 없습니다.

✅ 설치 후 새 터미널을 열고, 프로그램의 버전 확인 명령(예: --version)으로 정상 설치됐는지 확인하세요.

이 레포의 README에 적힌 실제 명령어를 그대로 가져왔습니다.

3. Python

쉬움

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Python 3 설치 시 'Add Python to PATH' 옵션을 꼭 체크하세요 (Windows).

pip install automatic-speech-recognition

PyPI에 배포된 패키지를 바로 설치합니다. 소스 클론이 필요 없습니다.

✅ 에러 메시지 없이 실행되고 터미널에 안내 문구가 출력되면 정상입니다.

이 레포의 README에 적힌 실제 명령어를 그대로 가져왔습니다.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

### Automatic Speech Recognition

The project aim is to distill the Automatic Speech Recognition research.
At the beginning, you can load a ready-to-use pipeline with a pre-trained model.
Benefit from the eager `TensorFlow 2.0` and freely monitor model weights, activations or gradients.

```python
import automatic_speech_recognition as asr

file = 'to/test/sample.wav'  # sample rate 16 kHz, and 16 bit depth
sample = asr.utils.read_audio(file)
pipeline = asr.load('deepspeech2', lang='en')
pipeline.model.summary()     # TensorFlow model
sentences = pipeline.predict([sample])
```

<br>

We support english (thanks to [Open Seq2Seq](https://nvidia.github.io/OpenSeq2Seq/html/speech-recognition.html#speech-recognition)).
The evaluation results of the English benchmark LibriSpeech dev-clean are in the table.
To reference, the DeepSpeech (Mozilla) achieves around 7.5% WER, whereas the state-of-the-art (RWTH Aachen University) equals 2.3% WER
(recent evaluation results can be found [here](https://paperswithcode.com/sota/speech-recognition-on-librispeech-test-clean)).
Both of them, use the external language model to boost results.
By comparison, _humans_ achieve 5.83% WER [here](https://arxiv.org/abs/1512.02595v1) (LibriSpeech dev-clean)

| Model Name    | Decoder | WER-dev |
| :---          |  :---:  |  :---:  |
| `deepspeech2` | greedy  |   6.71  |

<br>

Shortly it turns out that you need to adjust pipeline a little bit.
Take a look at the [CTC Pipeline](automatic_speech_recognition/pipeline/ctc_pipeline.py).
The pipeline is responsible for connecting a neural network model 
with all non-differential transformations (features extraction or prediction decoding).
Pipeline components are independent.
You can adjust them to your needs e.g. use more sophisticated feature extraction,
different data augmentation, or add the language model decoder (static n-grams or huge transformers).
You can do much more like distribute the training using the [Strategy](https://www.tensorflow.org/guide/distributed_training),
or experiment with [mixed precision](https://www.tensorflow.org/api_docs/python/tf/keras/mixed_precision/experimental/Policy) policy.

<br>

```python
import numpy as np
import tensorflow as tf
import automatic_speech_recognition as asr

dataset = asr.dataset.Audio.from_csv('train.csv', batch_size=32)
dev_dataset = asr.dataset.Audio.from_csv('dev.csv', batch_size=32)
alphabet = asr.text.Alphabet(lang='en')
features_extractor = asr.features.FilterBanks(
    features_num=160,
    winlen=0.02,
    winstep=0.01,
    winfunc=np.hanning
)
model = asr.model.get_deepspeech2(
    input_dim=160,
    output_dim=29,
    rnn_units=800,
    is_mixed_precision=False
)
optimizer = tf.optimizers.Adam(
    lr=1e-4,
    beta_1=0.9,
    beta_2=0.999,
    epsilon=1e-8
)
decoder = asr.decoder.GreedyDecoder()
pipeline = asr.pipeline.CTCPipeline(
    alphabet, features_extractor, model, optimizer, decoder
)
pipeline.fit(dataset, dev_dataset, epochs=25)
pipeline.save('/checkpoint')

test_dataset = asr.dataset.Audio.from_csv('test.csv')
wer, cer = asr.evaluate.calculate_error_rates(pipeline, test_dataset)
print(f'WER: {wer}   CER: {cer}')
```

<br>

#### Installation
You can use pip:
```bash
pip install automatic-speech-recognition
```
Otherwise clone the code and create a new environment via [conda](https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html#):
```bash
git clone https://github.com/rolczynski/Automatic-Speech-Recognition.git
conda env create -f=environment.yml     # or use: environment-gpu.yml
conda activate Automatic-Speech-Recognition
```

<br>

#### References

The fundamental repositories:
- Baidu - [DeepSpeech2 - A PaddlePaddle implementation of DeepSpeech2 architecture for ASR](https://github.com/PaddlePaddle/DeepSpeech)
- NVIDIA - [Toolkit for efficient experimentation with Speech Recognition, Text2Speech and NLP](https://nvidia.github.io/OpenSeq2Seq)
- RWTH Aachen University - [The RWTH extensible training framework for universal recurrent neural networks](https://github.com/rwth-i6/returnn)
- TensorFlow - [The implementation of DeepSpeech2 model](https://github.com/tensorflow/models/tree/master/research/deep_speech)
- Mozilla - [DeepSpeech - A TensorFlow implementation of Baidu's DeepSpeech architecture](https://github.com/mozilla/DeepSpeech) 
- Espnet - [End-to-End Speech Processing Toolkit](https://github.com/espnet/espnet)
- Sean Naren - [Speech Recognition using DeepSpeech2](https://github.com/SeanNaren/deepspeech.pytorch)

Moreover, you can explore the GitHub using key phrases like `ASR`, `DeepSpeech`, or `Speech-To-Text`.
The list [wer_are_we](https://github.com/syhw/wer_are_we), an attempt at tracking states of the art,
can be helpful too.