CoolFace
Modelpublic

OpenVoiceOS/parakeet-rnnt-120m-eou-coreml-6bit

sourceHugging Facecc-by-4.0updated 5mo agoView on Hugging Face
0likes6downloads
Model Card

parakeet-rnnt-120m-eou-coreml-6bit

CoreML conversion of nvidia/parakeet_realtime_eou_120m-v1 — 6BIT PALETTIZE KMEANS quantized.

ArchitectureRNNT
LanguageEnglish
Sample rate16000 Hz
Max audio15.0s
Vocab size1026
FrameworkNVIDIA NeMo → CoreML (coremltools)

Components

FileComponentBest compute
parakeet_mel_encoder.mlpackagemel_encoderANE / GPU
parakeet_decoder.mlpackagedecoderCPU only
parakeet_joint_decision_single_step.mlpackagejointdecisionsingle_stepANE / GPU

Usage

bash
pip install ovos-stt-plugin-coreml
python
from ovos_stt_plugin_coreml import CoremlSTT
from ovos_plugin_manager.utils.audio import AudioFile

stt = CoremlSTT(config={"metadata": "metadata.json"})

with AudioFile("speech.wav") as f:
    audio = f.read()
print(stt.execute(audio))

Source model

nvidia/parakeet_realtime_eou_120m-v1