Copy page

VHVoiceHub/asr_nemo

NeMoASR

Runs VoiceHub's native QuartzNet15x5 graph from the pinned NeMo/NGC source.

Automatic speech recognitionVoiceHub-nativenemo-asrParameters: Not reportedLanguage: enTraining: nativeLicense: NVIDIA-NGC-Terms

Parameter metadata: Not reported: the audited metadata available for the registered default does not provide an exact parameter total.

Use this model Copy model ID

Resources

Paper Upstream GitHub VoiceHub source

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 VoiceHub's native QuartzNet15x5 graph from the pinned NeMo/NGC source.

Inputs and controls: The audited release is English-only and supports CTC word timestamps, not arbitrary NeMo architectures.

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/nemo/stt_en_quartznet15x5',
    model_type='asr_nemo',
    device="cuda",
    lazy_load=True,
)
output = model.transcribe(
    AUDIO_FILE,
    language="en",
    return_timestamps="word",
)
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_nemo is a VoiceHub automatic speech recognition integration. This page is generated from its registry contract.

Property	Value
Task	Automatic speech recognition
Architecture	nemo-asr
Runtime	VoiceHub-native
Languages	en
Capabilities	automatic-speech-recognition, english, timestamps, safetensors, fine-tuning, voicehub-native, ctc
Reusable components	—
Normalized output	ASROutput

Language support

Supported language abbreviations

en

Paper and GitHub

• Paper: NeMo: a toolkit for building AI applications using Neural Modules
• Upstream GitHub: NVIDIA NeMo
• VoiceHub source: VoiceHub model implementation

Configuration

Load configuration without constructing the model:

from voicehub import AutoConfig

config = AutoConfig.for_model('asr_nemo')
print(config.model_type)

Property	Value
Canonical model type	asr_nemo
Configuration class	NeMoASRConfig
Architecture class	NeMoASRForSpeechRecognition

Processing

Create the registered processor without allocating model weights:

from voicehub import AutoProcessor

processor = AutoProcessor.from_pretrained(
    'nvidia/nemo/stt_en_quartznet15x5',
    model_type='asr_nemo',
)
print(type(processor).__name__)

Inference

The Usage example returns ASROutput through AutoModelForSpeechRecognition.

Input and output contract

Property	Value
Readiness	integrated-raw
Data architecture	ctc
Sample rate	16,000 Hz
Contract getter	get_asr_dataset_spec('asr_nemo')

Variant	Required fields	One of	Boundary	Other rules
raw-audio	audio	text / transcription / transcript	Source	at most one: text / transcription / transcript
nemo-ctc-waveform-model-ready	input_signal, input_signal_length, labels, label_lengths	—	Prepared	—
nemo-ctc-feature-model-ready	processed_signal, processed_signal_length, labels, label_lengths	—	Prepared	—

NeMo QuartzNet waveform and CTC 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	ctc
Recipe	single-phase
Default phase	speech_recognition
Training checkpoint	nvidia/nemo/stt_en_quartznet15x5
Native training graph	yes

Phase	Kind	Components	Required inputs	Loss keys
speech_recognition	objective	model	—	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/nemo/stt_en_quartznet15x5
Hugging Face ID	Not published / not applicable No canonical Hugging Face repository for the exact audited QuartzNet15x5 release; VoiceHub resolves the pinned NeMo/NGC artifact instead.
Checkpoint status	Pinned NeMo/NGC QuartzNet15x5 release; VoiceHub converts the exact audited graph into its safe native artifact boundary
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_nemo.NeMoASRForSpeechRecognition
Configuration	voicehub.models.asr_nemo.NeMoASRConfig
Source provenance	voicehub/architectures/nemo_ctc/SOURCE.json
License	NVIDIA-NGC-Terms

The QuartzNet checkpoint is governed by the NVIDIA NGC Terms of Use; the VoiceHub-owned architecture code is Apache-2.0. Commercial use: review required.

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

NeMoASRConfig

View source

NeMoASRConfig(**config_kwargs)

Parameters

• **config_kwargs — Configuration fields validated by NeMoASRConfig.

Model

NeMoASRForSpeechRecognition

View source

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

Parameters

• pretrained_model_name_or_path — Hub ID or compatible local directory.
• model_type — Canonical model type; use 'asr_nemo'.
• config — Optional preloaded NeMoASRConfig instance.
• **model_kwargs — Model-specific loading arguments.

from voicehub import get_model_spec

spec = get_model_spec('asr_nemo')
print(spec.display_name, spec.task.value)

Purpose	Public object
Discover	get_model_spec('asr_nemo')
Load and run	AutoModelForSpeechRecognition
Configure	NeMoASRConfig
Process	AutoProcessor
Model implementation	NeMoASRForSpeechRecognition
Normalized output	ASROutput
Training contract	get_training_spec('asr_nemo')
Optimization lifecycle	available_optimization_passes, apply_optimization_plan, optimization_manifest, restore_optimization_plan

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