25 Feb 2026
Efficient fine-tuning of whisper models for Amazigh ASR: leveraging LoRA and its variantsPublisher siteMultimedia Tools and Applications, Natural Language Processing Techniques
UniversitiesFadoua Ataa Allah
Active, last published 2026
In natural language processing techniques
Fadoua Ataa
Allah
Scholar“Research focused on Natural Language Processing Techniques and adjacent fields.”
Claim Your Profile7
H-index
152
Citations
Fadoua has not officially claimed their profile. If you are Fadoua, claim it free.
Claim Your Profile
Pioneer
LOCKED
Pathfinder
LOCKED
Scholar
EARNED
Innovator
LOCKED
Inventor
LOCKED
Catalyst
LOCKED
Expert
UPGRADE
01 · About the work
What I study
adoua Ataa Allah is a researcher working on Natural Language Processing Techniques. A fuller summary will appear once they update their profile.
02 · Reach
Publication output and reach
Fadoua's work has been cited 152 times across 4 active years. Bars show publications per year; the line shows cumulative citation totals across publications first appearing in each year.
03 · Research focus
Topics
Natural Language Processing Techniques×19Language, Linguistics, Cultural Analysis×12Speech Recognition and Synthesis×5Topic Modeling×4Handwritten Text Recognition Techniques×4
04 · Where I am
Affiliations
Affiliation details will appear once this profile is updated.
Research themes (UN SDGs)
05 · Selected work
Recent publications
9 works1 Jan 20251cites
Impact of Spectrogram Resizing for Automatic Speech Recognition Case of Amazigh Isolated WordPublisher siteCommunications in computer and information science, Speech Recognition and Synthesis
19 Dec 20242cites
Transformer-based Arabic language lemmatizationPublisher site1 Nov 20242cites
Optimizing Whisper models for Amazigh ASR: a comparative analysisPublisher siteInternational Journal of Speech Technology, Law, Economics, and Judicial Systems
21 Oct 20241cites
Microwave Sensors for Permittivity Measurement of Solid MaterialsPublisher site21 Oct 2024
Optimized Microwave Ring-Gap Resonator for Microchannel Particules State MonitoringPublisher site1 Jan 20243cites
Amazigh Speech Recognition via Parallel CNN Transformer-Encoder ModelPublisher siteLecture notes in information systems and organisation, Speech Recognition and Synthesis
1 Jan 20242cites
Enhancing Automatic Speech Recognition Systems for Amazigh Language Through Data AugmentationPublisher siteLecture notes in networks and systems, Speech Recognition and Synthesis
Expert Academic Profiles go further
Citation trend analytics, exportable publication records, and direct enquiry routing from industry and funders.
Are you Fadoua?
Claim it to verify your details, connect with industry partners, and explore Fractional Domain Expert opportunities through MARATTO™.
African researchers are full of potential that too often goes unrecognised. Your MARATTO™ profile gives your work the commercial pathway it deserves.
Claim Your Profile
Free to claim · takes a minute to verify it’s you