Designing a Smart Device for Speech Stammer Detection Features based on Artificial Neural Network

Stammering is a speech disorder characterized
by disruptions in verbal fluency, often affecting an individual’s
confidence and social interactions. This project introduces a
real-time stammering detection and correction system that
utilizes Natural Language Processing (NLP) and Artificial
Neural Networks (ANN) to enhance speech fluency. The system
functions by capturing live audio input, processing it through
NLP techniques to convert speech into text, and applying noise
reduction methods to improve clarity. Once the audio is
transcribed, an ANN-based algorithm is used to analyze the
spoken language, accurately identifying instances of
stammering. By leveraging machine learning, the system
effectively distinguishes between natural speech patterns and
stammered words or syllables, ensuring precise detection.
Upon detecting stammered speech, the system reconstructs it
into a more fluent form by eliminating unnecessary repetitions
and prolongations while preserving the original meaning and
intonation of the speaker. The refined speech is then
synthesized back into audio, ensuring smooth and natural
communication. This real-time process allows individuals with
stammering disorders to speak more confidently, as the system
provides immediate feedback and correction, significantly
reducing speech disruptions. The incorporation of NLP
enhances speech understanding, while ANN enables adaptive
learning, ensuring continuous improvement in detection
accuracy. Additionally, the integration of noise reduction
techniques ensures that external disturbances do not interfere
with speech processing, making the system more reliable in
diverse environments.