paper-with-me

홈 › Papers

Non-Autoregressive ASR with Self-Conditioned Folded Encoders

2022-02-17 · Tatsuya Komatsu

This paper proposes CTC-based non-autoregressive ASR with self-conditioned folded encoders. The proposed method realizes non-autoregressive ASR with fewer parameters by folding the conventional stack of encoders into only two blocks; base encoders and folded encoders. The base encoders convert the input audio features into a neural representation suitable for recognition. This is followed by the folded encoders applied repeatedly for further refinement. Applying the CTC loss to the outputs of all encoders enforces the consistency of the input-output relationship. Thus, folded encoders learn to perform the same operations as an encoder with deeper distinct layers. In experiments, we investigate how to set the number of layers and the number of iterations for the base and folded encoders. The results show that the proposed method achieves a performance comparable to that of the conventional method using only 38% as many parameters. Furthermore, it outperforms the conventional method when increasing the number of iterations.

📄 PDF Abstract BibTeX arXiv:2202.08474

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

CTC Loss 설명 없음
BASE 설명 없음

Similar Papers 제목 키워드 기반

HCNAF: Hyper-Conditioned Neural Autoregressive Flow and its Application for Probabilistic Occupancy Map Forecasting

2019-12-17 · CVPR 2020 6 · Geunseob Oh, Jean-Sebastien Valois

We introduce Hyper-Conditioned Neural Autoregressive Flow (HCNAF); a powerful universal distribution approximator designed to model arbitrarily complex conditional probability density functions. HCNAF consists of a neura…

Density Estimation

TESS: Text-to-Text Self-Conditioned Simplex Diffusion

2023-05-15 · Rabeeh Karimi Mahabadi, Hamish Ivison, Jaesung Tae, James Henderson 외

Diffusion models have emerged as a powerful paradigm for generation, obtaining strong performance in various continuous domains. However, applying continuous diffusion models to natural language remains challenging due t…

Natural Language UnderstandingParaphrase GenerationQuestion GenerationQuestion-Generation+2

ImageFolder: Autoregressive Image Generation with Folded Tokens

2024-10-02 · Xiang Li, Kai Qiu, Hao Chen, Jason Kuen 외

Image tokenizers are crucial for visual generative models, e.g., diffusion models (DMs) and autoregressive (AR) models, as they construct the latent representation for modeling. Increasing token length is a common approa…

Image GenerationImage ReconstructionQuantization

Test-Time Compute Scaling for ASR with Depth-Conditioned Looped Transformers

2026-06-03 · Yacouba Kaloga, Shashi Kumar, Shakeel A. Sheikh, Driss Khalil 외 arxiv

End-to-end ASR systems typically use fixed-depth acoustic encoders at inference, making it difficult to trade additional test-time computation for improved recognition without training a larger model. A natural approach …

Speech Recognition

Self-conditioned Embedding Diffusion for Text Generation

2022-11-08 · Robin Strudel, Corentin Tallec, Florent Altché, Yilun Du 외

Can continuous diffusion models bring the same performance breakthrough on natural language they did for image generation? To circumvent the discrete nature of text data, we can simply project tokens in a continuous spac…

Image GenerationLanguage ModelingLanguage ModellingText Generation