REVIEW 3 cited by
Efficient Streaming LLM for Speech Recognition
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original 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.
Forward citations
Cited by 3 Pith papers
-
Evaluation of LLMs in Speech is Often Flawed: Test Set Contamination in Large Language Models for Speech Recognition
LibriSpeech and Common Voice evaluation sentences leak into the Pile, and controlled LLM pretraining experiments show that contamination biases output probabilities even when error rates barely change.
-
SimulS2ST-Omni: Data-Efficient Streaming Speech-to-Speech Translation via Explicit Trajectory Supervision
A joint text-code trajectory supervision recipe lets a two-stream speech LM achieve competitive long-form streaming S2ST with ~2k hours of paired speech.
-
MFLA: Monotonic Finite Look-ahead Attention for Streaming Speech Recognition
MFLA adds finite look-ahead attention plus a CIF-based token counter to Whisper, enabling streaming recognition with a wait-k latency-quality trade-off.
Discussion (0). Continue with ORCID to comment.