Official Technical Resources: NVIDIA (page 6)
Official Technical Resources · NVIDIA: 264 entries in total; this page lists entries 76–90.
Data Acquisition Concepts
Core concepts for acquiring text data from remote sources including DocumentDownloader, DocumentIterator, and DocumentExtractor components
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
Text Data Curation Pipeline
Comprehensive overview of NeMo Curator's text curation pipeline architecture including data acquisition and processing
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
Text Processing Concepts
Text processing workflows including quality filtering, fuzzy deduplication, content cleaning, and pipeline design
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
Video Curation Concepts
Essential concepts for video data curation including distributed processing, pipeline stages, and execution modes
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
Create and Load Custom Audio Manifests
Create and load custom audio manifests in JSONL format for NeMo Curator audio processing
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
ALM Data Curation
ALM data curation stages for constructing and filtering training windows from diarized audio segments
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
ALM Data Builder
ALMDataBuilderStage reference for constructing training windows from diarized audio segments with quality filtering
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
ALM Overlap Filtering
ALMDataOverlapStage reference for removing overlapping training windows based on configurable thresholds
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
ASR Inference
Perform automatic speech recognition using NeMo Framework models with GPU acceleration and batch processing
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
Duration Filtering
Filter audio samples by duration ranges, speech rate metrics, and temporal characteristics for optimal dataset quality
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
Audio Quality Filtering
Compose preprocessing, VAD, band, MOS scoring, and speaker separation stages to extract clean single-speaker training segments from raw audio
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
Preprocessing Stages
Audio preprocessing stages — mono conversion, segment concatenation, and timestamp mapping for downstream filtering pipelines
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
SIGMOS Filter
Filter audio across seven independent perceptual quality dimensions — noise, overall, signal, coloration, discontinuity, loudness, reverb — using SIGMOSFilterStage
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
Speaker Separation
Diarize audio with offline SortFormer or streaming Sortformer and fan out one task per speaker for downstream per-speaker filtering
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25
UTMOS Filter
Filter audio segments by predicted Mean Opinion Score using UTMOSFilterStage and the utmos22_strong model
Evidence use: For understanding GPU computing, training, retrieval, inference serving, reliability or deployment decisions.
Limitations: Directly evidences only the implementation and claims of NVIDIA's technology stack. Recheck versioned interfaces and experimental capabilities before use.
Related concepts: Pre-training · Model Evaluation and Benchmarks · Multimodal Models
Link last checked: 2026-09-25