Reza2kn/visualears-fastconformer-fa-full-ab-onnx-fp

VisualEars FastConformer Persian ASR ONNX FP Export

Derivative export of Reza2kn/visualears-fastconformer-fa-full-ab.

Artifact

• Format: ONNX FP fixed-length acoustic CTC-core export
• Quantization/conversion: FP export reference; no weight quantization
• Runtime validation: ONNX Runtime CPU
• Validation result: 100.00% CTC argmax parity
• Size: 461 MB ONNX external-data pair

Validation

Runtime parity was checked against PyTorch CTC logits on 16 calibration clips padded to 2005 mel frames. The metric is CTC argmax token agreement versus the PyTorch reference logits, not end-to-end WER.

Usage Boundary

These are fixed-frame acoustic CTC-core exports. They take precomputed log-mel features as processed_signal; they are not full raw-audio-to-text pipelines by themselves.

Notes

Reference fixed-core ONNX export. Input is precomputed log-mel processed_signal, not raw audio.