No existing corpus simultaneously provides L2 speech, long-form presentation structure, accent labels, and prosodic annotations¶

Claim · All claims
Evidence · 1 study · review r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? A review of major corpora finds none combining all four properties needed for presentation-coaching research, limiting training and evaluation on discourse prosody and slide-aligned pacing. → Wen Liang 2026

Evidence¶

Wen Liang 2026¶

Wen Liang, Li Siyan, Zackary Rackauckas, and Julia Hirschberg. (2026). A Survey of Automated Presentation Coaching: Systems, Methods, and Open Challenges. https://arxiv.org/abs/2606.27380 doi:10.48550/arxiv.2606.27380

q2 · i? · review · r2

Survey assessment of major corpora (Table 2, Section 6) against four criteria: L2 speakers, long-form speech, accent/L1 labels, and prosody labels. The survey concludes that "no existing corpus simultane- ously provides" all four, noting TIMIT and L2-ARCTIC focus on isolated sentences while TED-LIUM 3 and GigaSpeech are long-form but not L2-specific.

"no existing corpus simultane- ously provides L2 speech, long-form presentation structure, accent labels, and prosodic annotations."

Discussion¶

Related Claims¶

• No surveyed presentation-coaching system covers all five taxonomy dimensions simultaneously — related
• CAPT software shows promising results for segmental pronunciation, while prosodic features and fluency still require further research and development — related