콘텐츠로 이동

nvidiaparakeet-tdt-0.6b-v3

ParakeetTDT

Runs the native Parakeet TDT decoder and returns its calibrated timestamp segments.

Automatic speech recognitionVoiceHub-nativeparakeet-tdtParameters: 627MLanguages: en, es +23Training: nativeLicense: CC-BY-4.0

Parameter metadata: Exact learned-parameter total for VoiceHub's audited native primary graph at the registered default selection; separately loaded auxiliary models are excluded.

Use this model Checkpoint
Resources
Upstream GitHub VoiceHub source Open in Colab

Usage

Complete the VoiceHub installation once, then run this repository-authored example. Model pages intentionally contain no package-install command.

This example is maintained against VoiceHub's public API; it is not copied from an upstream demo or package README.

Model-specific path: Runs the native Parakeet TDT decoder and returns its calibrated timestamp segments.

Inputs and controls: The registered multilingual release accepts automatic language handling; inspect the returned language metadata.

from pathlib import Path

from voicehub import AutoModelForSpeechRecognition

AUDIO_FILE = Path("speech.wav")
if not AUDIO_FILE.is_file():
    raise FileNotFoundError(AUDIO_FILE)

model = AutoModelForSpeechRecognition.from_pretrained(
    'nvidia/parakeet-tdt-0.6b-v3',
    model_type='asr_parakeet_tdt',
    device="cuda",
    lazy_load=True,
)
output = model.transcribe(
    AUDIO_FILE,
    return_timestamps=True,
)
print(output.text)
for segment in output.segments:
    print(segment.start, segment.end, segment.text, segment.confidence)

Use authorized recordings. Verify hardware needs and pin a revision in production.

Overview

asr_parakeet_tdt is a VoiceHub automatic speech recognition integration. This page is generated from its registry contract. Open the asr_parakeet_tdt Colab notebook.

Property Value
Task Automatic speech recognition
Architecture parakeet-tdt
Runtime VoiceHub-native
Languages en, es, fr, de, … complete audited list below
Capabilities automatic-speech-recognition, multilingual, timestamps, long-form, safetensors, fine-tuning, voicehub-native, native-runtime
Reusable components —
Normalized output ASROutput

Language support

Supported language abbreviations

en, es, fr, de, bg, hr, cs, da, nl, et, fi, el, hu, it, lv, lt, mt, pl, pt, ro, sk, sl, sv, ru, uk

Paper and GitHub

Configuration

Load configuration without constructing the model:

from voicehub import AutoConfig

config = AutoConfig.for_model('asr_parakeet_tdt')
print(config.model_type)
Property Value
Canonical model type asr_parakeet_tdt
Configuration class ParakeetTDTASRConfig
Architecture class ParakeetTDTForSpeechRecognition

Processing

Create the registered processor without allocating model weights:

from voicehub import AutoProcessor

processor = AutoProcessor.from_pretrained(
    'nvidia/parakeet-tdt-0.6b-v3',
    model_type='asr_parakeet_tdt',
)
print(type(processor).__name__)

Inference

The Usage example returns ASROutput through AutoModelForSpeechRecognition.

Input and output contract

Property Value
Readiness integrated-raw
Data architecture tdt
Sample rate 16,000 Hz
Contract getter get_asr_dataset_spec('asr_parakeet_tdt')
Variant Required fields One of Boundary Other rules
raw-audio audio text / transcription / transcript Source at most one: text / transcription / transcript
parakeet-tdt-model-ready input_features, attention_mask, labels, decoder_input_ids — Prepared —

Parakeet token-duration transducer audio and transcript records. See the data workflow.

Training and optimization

Use available_optimization_passes() to discover reversible public passes. Unsupported runtime or hardware fails closed before mutation.

Training contract

Property Value
Support native
Family tdt
Recipe single-phase
Default phase speech_recognition
Training checkpoint nvidia/parakeet-tdt-0.6b-v3
Native training graph yes
Phase Kind Components Required inputs Loss keys
speech_recognition objective model.encoder, model.encoder_projector, model.decoder, model.joint input_features, attention_mask, labels, decoder_input_ids loss

The integration accepts its declared source or prepared contract directly. Call model.validate_training_support() first, then follow the training workflow.

Checkpoints, provenance, license, and limitations

Property Value
Default checkpoint nvidia/parakeet-tdt-0.6b-v3
Hugging Face ID nvidia/parakeet-tdt-0.6b-v3
Repository availability verified through the Hugging Face model API on 2026-08-11; pin a revision before production use.
Checkpoint status Registry default; pin an immutable revision for production and reproducible evidence
Optional dependency extra Core package
Hardware and runtime Usage selects cuda; verify checkpoint-specific requirements
Real-checkpoint evidence Release evidence; a registry default alone is not execution evidence
Implementation voicehub.models.asr_parakeet_tdt.modeling_asr_parakeet_tdt.ParakeetTDTForSpeechRecognition
Configuration voicehub.models.asr_parakeet_tdt.configuration_asr_parakeet_tdt.ParakeetTDTASRConfig
Source provenance voicehub/architectures/parakeet_tdt/SOURCE.json
License CC-BY-4.0

The pinned Parakeet TDT checkpoint and derivatives require CC-BY-4.0 attribution. The VoiceHub-owned architecture port is audited against Apache-2.0 Transformers and NeMo source. Commercial use: allowed by the registered terms.

Confirm the checkpoint revision, access terms, provenance, and license.

Limitations

  • No integration-specific checkpoint limitation is registered. Verify the selected checkpoint revision and its documented runtime requirements.
  • Validate memory, precision, and optional dependencies on the target system.
  • Public optimizations fail closed when the runtime or hardware cannot satisfy their validation contract; an unavailable pass is not reported as applied.
  • Contract tests do not replace the linked released-checkpoint evidence.

Public API

Use the stable configuration, processor, and task-model facades below.

Configuration

ParakeetTDTASRConfig

View source

ParakeetTDTASRConfig(**config_kwargs)

Parameters

  • **config_kwargs — Configuration fields validated by ParakeetTDTASRConfig.

Model

ParakeetTDTForSpeechRecognition

View source

AutoModelForSpeechRecognition.from_pretrained(
    pretrained_model_name_or_path,
    *,
    model_type='asr_parakeet_tdt',
    config=None,
    **model_kwargs,
)

Parameters

  • pretrained_model_name_or_path — Hub ID or compatible local directory.
  • model_type — Canonical model type; use 'asr_parakeet_tdt'.
  • config — Optional preloaded ParakeetTDTASRConfig instance.
  • **model_kwargs — Model-specific loading arguments.
from voicehub import get_model_spec

spec = get_model_spec('asr_parakeet_tdt')
print(spec.display_name, spec.task.value)
Purpose Public object
Discover get_model_spec('asr_parakeet_tdt')
Load and run AutoModelForSpeechRecognition
Configure ParakeetTDTASRConfig
Process AutoProcessor
Model implementation ParakeetTDTForSpeechRecognition
Normalized output ASROutput
Training contract get_training_spec('asr_parakeet_tdt')
Optimization lifecycle available_optimization_passes, apply_optimization_plan, optimization_manifest, restore_optimization_plan

See all model guides, inference, and the training matrix.