kshitijbhandari/Speech-Detection----Spoken-Digit
Spoken digit recognition (0–9) using LSTM networks. Compares raw waveforms vs. log-mel spectrograms with and without data augmentation. Best model achieves 95.5% F1 using augmented spectrograms. Built with TensorFlow, Librosa, and the Free Spoken Digit Dataset.
Stars
—
Forks
—
Language
Jupyter Notebook
License
—
Category
Last pushed
Mar 06, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/kshitijbhandari/Speech-Detection----Spoken-Digit"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
chatopera/efaqa-corpus-zh
❤️Emotional First Aid Dataset, 心理咨询问答、聊天机器人语料库
ikegami-yukino/pymlask
Emotion analyzer for Japanese text
declare-lab/conv-emotion
This repo contains implementation of different architectures for emotion recognition in conversations.
sarnthil/unify-emotion-datasets
A Survey and Experiments on Annotated Corpora for Emotion Classification in Text
kanchitank/Text-Emotion-Analysis
Automate detection of different emotions from paragraphs and predict overall emotion.