Fact · Speech recognition
414.63times faster than real time
In the Galvez et al. 2024 paper, the baseline decoder for Parakeet RNN-T 1.1B ran at an RTFx of 414.63 on an A100-80GiB PCIe GPU at batch size 32, meaning 414.63 times faster than real time.
Quote found word for word on the live source, last checked October 19, 2026
In the source's own words
“Parakeet RNN-T 1.1B Baseline 414.63 67.39 2.7 1 1 ... For all results, we use an A100-80GiB PCIe card, and a batch size of 32”
Speed of Light Exact Greedy Decoding for RNN-T Speech Recognition Models on GPU, Daniel Galvez et al., Proceedings of Interspeech 2024 (arXiv:2406.03791), 2024, Section 4 (Results) and Table 1, Parakeet RNN-T 1.1B Baseline row.
Open the original (DOI 10.48550/arXiv.2406.03791)- What it measures
- The paper defines RTFx as the number of hours an ASR model can transcribe in one hour of wall-clock time. It is an offline inference task on LibriSpeech test-other (5.1 hours), with bfloat16 inference. The Table 1 RTFx column is the overall RTFx, and the baseline row is the reference for the paper's speed-up factors.
- What it does not show
- Offline and batched at batch size 32, with bfloat16. This is not a batch-size-1 or single-file speed, so it is not directly comparable with the batch-1 A100 figure from the MDM-ASR paper. The model is NVIDIA's open Parakeet checkpoint; the row is a decoder measurement from a research paper.
In the data
A row of Speech Recognition Speed: Real-Time Factor Data, as Galvez et al. 2024, RNN-T baseline: 415 times faster than real time.
- System
- Parakeet RNN-T 1.1B, baseline decoder
- Kind
- Research system
- Hardware
- NVIDIA A100-80GiB PCIe GPU
- Test set
- LibriSpeech test-other, offline, batch size 32
- Year
- 2024