Murmur 提出分层分块方法,实现高效长上下文 ASR 推理
来源标题:Murmur: Efficient Long-Context ASR Inference with Hierarchical Chunking
Murmur 是一种用于长上下文语音识别(ASR)的高效推理方法,采用分层分块策略处理长音频输入。该方法旨在降低长上下文 ASR 推理的开销,具体性能数据尚未披露。
以下正文同步自 Hacker News 最新,版权归原站所有,已转换为易读排版。
TL;DR: Murmur speeds up long-context ASR models on audio up to 1 hour by 2.5× on average (up to 2.9×), end to end including diarization. It also lowers speaker-attributed error (tcpWER) compared with WhisperX by 7-17 points on English meetings and 10-51 points on Mandarin.
Single pass, Murmur, and WhisperX transcribing the same 32-minute AMI meeting on one H100, replayed at 8×.
Recent long-context ASR models, such as Microsoft VibeVoice-ASR and MOSS-Transcribe-Diarize, process an hour or more of audio in a single pass. They generate text, timestamps and speaker labels at the same time. An hour of audio is tens of thousands of audio tokens (about 27,000 for VibeVoice-ASR), so a single pass is slow: on one H100, VibeVoice-ASR takes about 2.5 minutes to transcribe a 44-minute meeting. These models can also fall into a repetition loop, which makes the whole request fail.
The other common approach is to split the recording into short chunks, such as the 30 s chunks WhisperX uses for Whisper. Timestamps and speaker labels then have to be stitched together across hundreds of chunk boundaries, which lowers accuracy, especially in speaker attribution. This design was built for short-context models, while long-context models can take in minutes of audio at once.
Neither option fits long-context models well. We revisit how long audio should be chunked and served for these models, and design Murmur, an inference system that closes the gap between the two approaches.
We first study how chunk size affects long-context ASR. Across 4 datasets and 2 languages (English and Mandarin), error drops quickly as chunks grow from 30 s and levels off within a few minutes (Figure 1). At 330 s, chunked decoding already improves on or matches single pass on word error rate (WER) and time-constrained word error rate (tcpWER) in most cases, so the model does not need the whole recording in one pass.
正文由 探How 从来源站点 RSS 同步,内容未经改写;遇到排版缺失或需要图片、视频时请以原文为准。