scionoftech/DeepAsr
Keras(Tensorflow) implementations of Automatic Speech Recognition
Implements end-to-end ASR using CTC loss with support for Baidu's DeepSpeech2 and custom architectures, featuring CuDNN-accelerated LSTMs, multi-GPU distributed training, and mixed precision capabilities. Includes modular pipeline components for audio feature extraction (MFCC/spectrogram), greedy and beam-search decoders with language model integration, and on-the-fly data generation for large datasets. Targets 16kHz mono WAV/FLAC inputs and provides pre-trained models for immediate inference or fine-tuning.
No commits in the last 6 months. Available on PyPI.
Stars
24
Forks
11
Language
Jupyter Notebook
License
AGPL-3.0
Category
Last pushed
Jan 13, 2022
Monthly downloads
683
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/scionoftech/DeepAsr"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
githubharald/CTCWordBeamSearch
Connectionist Temporal Classification (CTC) decoder with dictionary and language model.
githubharald/CTCDecoder
Connectionist Temporal Classification (CTC) decoding algorithms: best path, beam search, lexicon...
nl8590687/ASRT_SpeechRecognition
A Deep-Learning-Based Chinese Speech Recognition System 基于深度学习的中文语音识别系统
athena-team/athena
an open-source implementation of sequence-to-sequence based speech processing engine
hirofumi0810/tensorflow_end2end_speech_recognition
End-to-End speech recognition implementation base on TensorFlow (CTC, Attention, and MTL training)