parakeet-tdt-ctc-110m-coreml
CoreML conversion of nvidia/parakeet-tdt_ctc-110m.
| Architecture | TDT |
| Language | English |
| Sample rate | 16000 Hz |
| Max audio | 15.0s |
| Vocab size | 1024 |
| Framework | NVIDIA NeMo → CoreML (coremltools) |
Components
| File | Component | Best compute |
|---|---|---|
parakeet_mel_encoder.mlpackage |
mel_encoder | ANE / GPU |
parakeet_ctc_decoder.mlpackage |
ctc_decoder | ANE / GPU |
parakeet_decoder.mlpackage |
decoder | CPU only |
parakeet_joint_decision_single_step.mlpackage |
joint_decision_single_step | ANE / GPU |
Usage
pip install ovos-stt-plugin-coreml
from ovos_stt_plugin_coreml import CoremlSTT
stt = CoremlSTT(config={"metadata": "metadata.json"})
Source model
- Downloads last month
- -
Model tree for OpenVoiceOS/parakeet-tdt-ctc-110m-coreml
Base model
nvidia/parakeet-tdt_ctc-110m