# Timestamped Embedding-Matching Acoustic-to-Word CTC ASR

arXiv: https://arxiv.org/abs/2306.11473 · HTML: https://mlsanity.com/paper/2306.11473/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 38 references that are in the map.
- First posted on arXiv 20 Jun 2023.

## Builds on (most cited first, 12 of 12)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [Recent Advances in End-to-End Automatic Speech Recognition](https://mlsanity.com/paper/2111.01690/) (2021)
- [Neural Speech Recognizer: Acoustic-to-Word LSTM Model for Large Vocabulary Speech Recognition](https://mlsanity.com/paper/1610.09975/) (2016)
- [First-Pass Large Vocabulary Continuous Speech Recognition using Bi-Directional Recurrent DNNs](https://mlsanity.com/paper/1408.2873/) (2014)
- [Multi-view Recurrent Neural Acoustic Word Embeddings](https://mlsanity.com/paper/1611.04496/) (2016)
- [Acoustically Grounded Word Embeddings for Improved Acoustics-to-Word Speech Recognition](https://mlsanity.com/paper/1903.12306/) (2019)
- [SNDCNN: Self-normalizing deep CNNs with scaled exponential linear units for speech recognition](https://mlsanity.com/paper/1910.01992/) (2019)
- [Minimum Latency Training of Sequence Transducers for Streaming End-to-End Speech Recognition](https://mlsanity.com/paper/2211.02333/) (2022)
- [On Modeling ASR Word Confidence](https://mlsanity.com/paper/1907.09636/) (2019)
- [Acoustic Neighbor Embeddings](https://mlsanity.com/paper/2007.10329/) (2020)
- [Improvements to Embedding-Matching Acoustic-to-Word ASR Using Multiple-Hypothesis Pronunciation-Based Embeddings](https://mlsanity.com/paper/2210.16726/) (2022)

## Abstract (arXiv)

In this work, we describe a novel method of training an embedding-matching word-level connectionist temporal classification (CTC) automatic speech recognizer (ASR) such that it directly produces word start times and durations, required by many real-world applications, in addition to the transcription. The word timestamps enable the ASR to output word segmentations and word confusion networks without relying on a secondary model or forced alignment process when testing. Our proposed system has similar word segmentation accuracy as a hybrid DNN-HMM (Deep Neural Network-Hidden Markov Model) system, with less than 3ms difference in mean absolute error in word start times on TIMIT data. At the same time, we observed less than 5% relative increase in the word error rate compared to the non-timestamped system when using the same audio training data and nearly identical model size. We also contribute more rigorous analysis of multiple-hypothesis embedding-matching ASR in general.