ASR-command v2
Single-speaker English and German utterances. Each row contains its own WAV bytes and its final v2 target. Movement operands are continuous, time_ms or distance_mm. The original audio is reused byte for byte. Only additions require TTS. Source provenance, the text label map and batch completion counts are in _build/.
- Downloads last month
- 1,811