From speech to letters - using a novel neural network architecture for grapheme based ASR

By Florian Eyben, Martin Wöllmer, Björn Schuller, Alex Graves. 2009

52 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Signal Processing

Citations in the corpus

Fast and accurate recurrent neural network acoustic models for speech recognition (2015, background): “These have been shown to perform better than shallow models for speech recognition [7, 8, 9, 3].”
Exploring architectures, data and units for streaming end-to-end speech recognition with RNN-transducer (2017, methodology): “CTC has been widely used in previous works to train end-toend ASR models [8, 18, 19].”
Analyzing Hidden Representations in End-to-End Automatic Speech Recognition Systems (2017, positive): “The CTC loss then marginalizes over all possible sequences of symbols given a transcription.”
Hybrid Autoregressive Transducer (HAT) (2020, methodology): “The letter models [19, 20, 21] and word model [22] are examples of this neural model class.”
Understanding LSTM - a tutorial into Long Short-Term Memory Recurrent Neural Networks (2019, unknown): “A variant named BLSTM-CTC [28, 19, 17] ﬁnally outperformed HMMs, with recent improvements documented in [44, 77].”
On internal language representations in deep learning: an analysis of machine translation and speech recognition (2018, methodology): “The CTC objective function [131] marginalizes over all possible sequences of symbols given a transcription: ∑︁ 𝑖 log 𝑝(𝑙(𝑖)|𝑥(𝑖)) (1.14) where the probability of a label sequence 𝑙 given an input sequence 𝑥 is defined as: 𝑝(𝑙|𝑥) = ∑︁ 𝜋∈ℬ−1(𝑙) 𝑝(𝜋|𝑥) = ∑︁ 𝜋∈ℬ−1(𝑙) |𝑙(𝑖)|∏︁ 𝑗=1 𝜑”
Towards End-To-End Speech Recognition with Recurrent Neural Networks (2014, background): “The combination of bidirectional LSTM and CTC has been applied to character-level speech recognition before (Eyben et al., 2009), however the relatively shallow architecture used in that work did not deliver compelling results (the best character error rate was almost 20%).”