Low-resource languages remain underrepresented in ASR benchmarks, limiting the ability to reliably assess and compare model performance. We present a large-scale benchmark for automatic speech recognition (ASR) on low-resource languages in Africa. The benchmark harmonizes six public corpora into a unified, reproducible evaluation framework and evaluates 52 state-of-the-art ASR models spanning three model architectural paradigms across 39 African languages on diverse datasets. Models are evaluated using standard accuracy and efficiency metrics, word error rate (WER), character error rate (CER), and inverse real-time factor (RTFx). This benchmark introduces architecture-aware comparison that systematically analyzes both individual models and broader model families. In addition, we quantify linguistic variance across language families, revealing how these factors shape ASR performance across diverse African languages. We also provide a comprehensive accuracy and efficiency trade-off analysis directly relevant to deployment in low-resource settings. Together, the benchmark, evaluation pipeline, and results support transparent, reproducible, and deployment-aware evaluation of ASR models for low-resource languages.<\/p>\n","protected":false},"excerpt":{"rendered":"
Low-resource languages remain underrepresented in ASR benchmarks, limiting the ability to reliably assess and compare model performance. We present a large-scale benchmark for automatic speech recognition (ASR) on low-resource languages in Africa. The benchmark harmonizes six public corpora into a unified, reproducible evaluation framework and evaluates 52 state-of-the-art ASR models spanning three model architectural paradigms […]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr-author-ordering":[{"type":"user_nicename","value":"Mercy Muchai","user_id":"40846"},{"type":"text","value":"Nick Mumero Mwangi","user_id":0},{"type":"text","value":"Kevin Chege","user_id":0},{"type":"user_nicename","value":"Samuel Chege Maina","user_id":"40321"}],"msr_publishername":"Indaba Proceedings IJCAI 2026","msr_publisher_other":"","msr_booktitle":"","msr_chapter":"","msr_edition":"","msr_editors":"","msr_how_published":"","msr_isbn":"","msr_issue":"","msr_journal":"","msr_number":"","msr_organization":"Deep Learning Indaba","msr_pages_string":"","msr_page_range_start":"","msr_page_range_end":"","msr_series":"","msr_volume":"","msr_copyright":"","msr_conference_name":"Deep Learning Indaba 2026","msr_doi":"","msr_arxiv_id":"","msr_mag_id":"","msr_other_authors":"","msr_other_contributors":"","msr_speaker":"","msr_award":"","msr_affiliation":"","msr_institution":"","msr_host":"","msr_version":"","msr_duration":"","msr_release_tracker_id":"","msr_highlight_type":"","msr_date_display_format":"","msr_main_download_label":"","msr_external_link_label":"","msr_doi_label":"","msr_published_date":"2026-08","msr_startdate":"","msr_presentation_date":"","msr_highlight_text":"","msr_notes":"","msr_longbiography":"","msr_publicationurl":"","msr_external_url":"","msr_secondary_video_url":"","msr_conference_url":"https:\/\/deeplearningindaba.com\/2026\/","msr_journal_url":"","msr_year":2026,"msr_month":8,"msr_day":0,"msr_microsoftintellectualproperty":false,"msr_pub_id":"","msr_publication_uploader":[{"type":"file","title":"PazaBench-Large-Scale-ASR-Benchmark-for-Low-Resource-Languages-in-Africa.pdf","label_id":243109,"id":1188619,"viewUrl":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-content\/uploads\/2026\/10\/PazaBench-Large-Scale-ASR-Benchmark-for-Low-Resource-Languages-in-Africa.pdf"}],"msr_related_uploader":[{"type":"url","title":"https:\/\/huggingface.co\/spaces\/microsoft\/paza-bench","label_id":264520,"id":false,"viewUrl":false}],"msr_original_fields_of_study":[],"msr_s2_paper_id":"","msr_s2_pdf_url":"","msr_citation_count_updated":"","msr_citation_count":0,"msr_influential_citations":0,"msr_reference_count":0,"msr_s2_open_access":false,"msr_s2_author_ids":[],"msr_pub_ids":[],"msr_hide_image_in_river":0,"footnotes":""},"msr-research-highlight":[],"research-area":[13556],"msr-publication-type":[193716],"msr-publisher":[],"msr-publication-cta":[],"msr-focus-area":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[260464,247753],"msr-conference":[270640],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-1188617","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-artificial-intelligence","msr-locale-en_us","msr-field-of-study-natural-langage-processing","msr-field-of-study-speech-recognition"],"msr_publishername":"Indaba Proceedings IJCAI 2026","msr_edition":"","msr_affiliation":"","msr_published_date":"2026-08","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"Deep Learning Indaba","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":0,"msr_main_download":"","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"file","title":"PazaBench-Large-Scale-ASR-Benchmark-for-Low-Resource-Languages-in-Africa.pdf","label_id":243109,"id":1188619,"viewUrl":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-content\/uploads\/2026\/10\/PazaBench-Large-Scale-ASR-Benchmark-for-Low-Resource-Languages-in-Africa.pdf"}],"msr_related_uploader":[{"type":"url","title":"https:\/\/huggingface.co\/spaces\/microsoft\/paza-bench","label_id":264520,"id":false,"viewUrl":false}],"msr_citation_count":0,"msr_citation_count_updated":"","msr_s2_paper_id":"","msr_influential_citations":0,"msr_reference_count":0,"msr_arxiv_id":"","msr_s2_author_ids":[],"msr_s2_open_access":false,"msr_s2_pdf_url":null,"msr_attachments":[{"id":1188619,"url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-content\/uploads\/2026\/10\/PazaBench-Large-Scale-ASR-Benchmark-for-Low-Resource-Languages-in-Africa.pdf"},{"id":1188618,"url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-content\/uploads\/2026\/10\/PazaBench-Indaba-2026-Camera-Ready-Paper-Submission.pdf"}],"msr-author-ordering":[{"type":"user_nicename","value":"Mercy Muchai","user_id":40846,"rest_url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Mercy Muchai"},{"type":"text","value":"Nick Mumero Mwangi","user_id":0,"rest_url":false},{"type":"text","value":"Kevin Chege","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Samuel Chege Maina","user_id":40321,"rest_url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Samuel Chege Maina"}],"msr_impact_theme":[],"msr_research_lab":[1021599],"msr_event":[],"msr_group":[],"msr_project":[1119384],"publication":[],"video":[],"msr-tool":[],"msr_publication_type":"inproceedings","related_content":{"projects":[{"ID":1119384,"post_title":"Project Gecko","post_name":"project-gecko","post_type":"msr-project","post_date":"2025-02-20 05:34:34","post_modified":"2026-07-01 15:12:32","post_status":"publish","permalink":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/project\/project-gecko\/","post_excerpt":"A cross-lab initiative building equitable, trusted multilingual Copilots for population-scale use, starting with speech-first agricultural assistants for smallholder farmers in East Africa and South Asia","_links":{"self":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/1119384"}]}}]},"_links":{"self":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1188617","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":6,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1188617\/revisions"}],"predecessor-version":[{"id":1188626,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1188617\/revisions\/1188626"}],"wp:attachment":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1188617"}],"wp:term":[{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=1188617"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1188617"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=1188617"},{"taxonomy":"msr-publisher","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publisher?post=1188617"},{"taxonomy":"msr-publication-cta","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-cta?post=1188617"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=1188617"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1188617"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1188617"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=1188617"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=1188617"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=1188617"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1188617"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=1188617"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}