AmourWaltz/BayesLMs
Project of IEEE/ACM TASLP “Bayesian Neural Network Language Modeling for Speech Recognition”
This project helps speech recognition engineers improve the accuracy and robustness of their language models. By inputting existing neural network language models (like LSTM or Transformer) and speech data, it outputs enhanced Bayesian versions that better handle uncertainty. The primary users are researchers and practitioners working on advanced speech recognition systems.
No commits in the last 6 months.
Use this if you are a speech recognition engineer looking to implement or experiment with Bayesian neural network language models to improve your system's performance, particularly in terms of uncertainty handling.
Not ideal if you are a general machine learning practitioner seeking an off-the-shelf solution for common NLP tasks, as this is highly specialized for speech recognition language modeling research.
Stars
9
Forks
2
Language
Jupyter Notebook
License
—
Category
Last pushed
Jan 22, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/AmourWaltz/BayesLMs"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
julius-speech/julius
Open-Source Large Vocabulary Continuous Speech Recognition Engine
rolczynski/Automatic-Speech-Recognition
🎧 Automatic Speech Recognition: DeepSpeech & Seq2Seq (TensorFlow)
tabahi/formantfeatures
Extract frequency, power, width and dissonance of formants from wav files
libdriver/ld3320
LD3320 full-featured driver library for general-purpose MCU and Linux.
awsaf49/audio_classification_models
Tensorflow Audio Classification Models