Script-Native ASR for N'Ko: Anticipatory Transformer CTC Decoding and the CER Anchor

This paper preserves the technical ASR center of the \nko{} research program: an archived script-native trajectory checkpoint reporting \anchorcer{} character error rate on a \corpusn{}-pair Bambara corpus snapshot. The model uses frozen Whisper large-v3 acoustic features, a trainable Transformer CTC decoder, and a compact trajectory state that biases attention with speech-dynamic information. The result is the strongest retained ASR artifact in the project and is the correct way to discuss the phrase ``20 CER'' publicly. The paper gives the architecture, training regime, artifact contract, and claim boundaries. Input audio is encoded by Whisper large-v3 \citep{radford2023robust}; 1280-dimensional features are projected to a 768-dimensional decoder space, temporally downsampled, and decoded by a six-layer Transformer CTC head \citep{graves2006connectionist}. A trajectory module estimates a seven-dimensional state for each timestep: commitment, uncertainty, transition pressure, recovery margin, phase stiffness, novelty, and stability. This state produces an additive attention-logit bias before CTC emission, giving the decoder an anticipatory geometry over speech dynamics. The archived anchor was trained on \corpusn{} paired examples split into 232,476 training rows, 29,060 validation rows, and 29,060 test rows, with learning rate 0.0003, batch size 32, dropout 0.1, seed 42, and best validation loss 0.6358872798606507. The reported test CER is \anchorcer{}, computed as 216,225 edits over 1,050,967 reference characters. It is an archived checkpoint result with preserved metadata. Later low-learning-rate runs around 31\% CER are not comparable to the anchor because they used a different learning-rate regime. The conclusion is therefore bounded: direct \nko{} ASR reached a meaningful error regime under recorded settings, and the artifact should not be silently replaced by non-comparable runs.