项目作者: usimarit

项目描述 :
A Sweet Sound Event Localization and Detection like Tiramisu Cake in Tensorflow 2
高级语言: Python
项目地址: git://github.com/usimarit/TiramisuSELD.git
创建时间: 2020-09-07T15:49:33Z
项目社区:https://github.com/usimarit/TiramisuSELD

开源协议:Apache License 2.0

下载


TiramisuSELD :cake:




GitHub

python
tensorflow
ubuntu




Sound Event Localization and Detection in Tensorflow 2





TiramisuSELD implements some speech event localization and detection architectures.

Requirements

  • Python 3.6+
  • Tensorflow 2.2+: pip install tensorflow

Setup Environment and Datasets

Install tensorflow: pip3 install tensorflow or pip3 install tf-nightly (for using tflite)

Install packages: python3 setup.py install

  • To enable XLA, run TF_XLA_FLAGS=--tf_xla_auto_jit=2 $python_train_script

Clean up: python3 setup.py clean --all (this will remove /build contents)

Training & Testing

Example YAML Config Structure

  1. speech_config: ...
  2. model_config: ...
  3. decoder_config: ...
  4. learning_config:
  5. augmentations: ...
  6. dataset_config:
  7. train_paths: ...
  8. eval_paths: ...
  9. test_paths: ...
  10. tfrecords_dir: ...
  11. optimizer_config: ...
  12. running_config:
  13. batch_size: 8
  14. num_epochs: 20
  15. outdir: ...
  16. log_interval_steps: 500

See examples for some predefined ASR models.

References & Credits

  1. https://github.com/pquochuy/dcase2020-seld