Efficient Streaming LLM for Speech Recognition

• URL: http://arxiv.org/abs/2410.03752v1
• Date: Wed, 2 Oct 2024 01:54:35 GMT
• Title: Efficient Streaming LLM for Speech Recognition
• Authors: Junteng Jia, Gil Keren, Wei Zhou, Egor Lakomkin, Xiaohui Zhang, Chunyang Wu, Frank Seide, Jay Mahadeokar, Ozlem Kalinli,
• Abstract summary: SpeechLLM-XL is a linear scaling decoder-only model for streaming speech recognition. It achieves no quality degradation on long form utterances 10x longer than the training utterances.
• Score: 23.151980358518102
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent works have shown that prompting large language models with audio encodings can unlock speech recognition capabilities. However, existing techniques do not scale efficiently, especially while handling long form streaming audio inputs -- not only do they extrapolate poorly beyond the audio length seen during training, but they are also computationally inefficient due to the quadratic cost of attention. In this work, we introduce SpeechLLM-XL, a linear scaling decoder-only model for streaming speech recognition. We process audios in configurable chunks using limited attention window for reduced computation, and the text tokens for each audio chunk are generated auto-regressively until an EOS is predicted. During training, the transcript is segmented into chunks, using a CTC forced alignment estimated from encoder output. SpeechLLM-XL with 1.28 seconds chunk size achieves 2.7%/6.7% WER on LibriSpeech test clean/other, and it shows no quality degradation on long form utterances 10x longer than the training utterances.

Related papers

• SyllableLM: Learning Coarse Semantic Units for Speech Language Models [21.762112843104028]
  We introduce a controllable self-supervised technique to merge speech representations into coarser syllable-like units. Our method produces controllable-rate semantic units at as low as 5Hz and 60bps and SotA inc segmentation and clustering. SyllableLM achieves significant improvements in efficiency with a 30x reduction in training compute and a 4x wall-clock inference speedup.
  arXiv  Detail & Related papers  (2024-10-05T04:29:55Z)
• Low Frame-rate Speech Codec: a Codec Designed for Fast High-quality Speech LLM Training and Inference [10.909997817643905]
  We present the Low Frame-rate Speech Codec (LFSC): a neural audio that leverages a finite scalar quantization and adversarial training with large speech language models to achieve high-quality audio compression with a 1.89 kbps and 21.5 frames per second. We demonstrate that our novel LLM can make the inference of text-to-speech models around three times faster while improving intelligibility and producing quality comparable to previous models.
  arXiv  Detail & Related papers  (2024-09-18T16:39:10Z)
• XLAVS-R: Cross-Lingual Audio-Visual Speech Representation Learning for Noise-Robust Speech Perception [62.660135152900615]
  Speech recognition and translation systems perform poorly on noisy inputs. XLAVS-R is a cross-lingual audio-visual speech representation model for noise-robust speech recognition and translation.
  arXiv  Detail & Related papers  (2024-03-21T13:52:17Z)
• LauraGPT: Listen, Attend, Understand, and Regenerate Audio with GPT [65.69648099999439]
  Generative Pre-trained Transformer (GPT) models have achieved remarkable performance on various natural language processing tasks. We propose LauraGPT, a novel unified audio-and-text GPT-based LLM for audio recognition, understanding, and generation.
  arXiv  Detail & Related papers  (2023-10-07T03:17:59Z)
• Large-scale unsupervised audio pre-training for video-to-speech synthesis [64.86087257004883]
  Video-to-speech synthesis is the task of reconstructing the speech signal from a silent video of a speaker. In this paper we propose to train encoder-decoder models on more than 3,500 hours of audio data at 24kHz. We then use the pre-trained decoders to initialize the audio decoders for the video-to-speech synthesis task.
  arXiv  Detail & Related papers  (2023-06-27T13:31:33Z)
• AudioPaLM: A Large Language Model That Can Speak and Listen [79.44757696533709]
  We introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models. It can process and generate text and speech with applications including speech recognition and speech-to-speech translation.
  arXiv  Detail & Related papers  (2023-06-22T14:37:54Z)
• Exploring the Role of Audio in Video Captioning [59.679122191706426]
  We present an audio-visual framework, which aims to fully exploit the potential of the audio modality for captioning. We propose new local-global fusion mechanisms to improve information exchange across audio and video.
  arXiv  Detail & Related papers  (2023-06-21T20:54:52Z)
• Pre-Training Transformer Decoder for End-to-End ASR Model with Unpaired Speech Data [145.95460945321253]
  We introduce two pre-training tasks for the encoder-decoder network using acoustic units, i.e., pseudo codes. The proposed Speech2C can relatively reduce the word error rate (WER) by 19.2% over the method without decoder pre-training.
  arXiv  Detail & Related papers  (2022-03-31T15:33:56Z)
• Attentional Speech Recognition Models Misbehave on Out-of-domain Utterances [16.639133822656458]
  We decode audio from the British National Corpus with an attentional encoder-decoder model trained solely on the LibriSpeech corpus. We observe that there are many 5-second recordings that produce more than 500 characters of decoding output. A frame-synchronous hybrid (DNN-HMM) model trained on the same data does not produce these unusually long transcripts.
  arXiv  Detail & Related papers  (2020-02-12T18:53:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.