arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2303.00747cs.SDeess.AS
收藏

WhisperX:长音频的时间精确语音转录

WhisperX: Time-Accurate Speech Transcription of Long-Form Audio

University of Oxford(牛津大学)

Max Bain, Jaesung Huh, Tengda Han, Andrew Zisserman

AI总结:

WhisperX 通过语音活动检测与强制音素对齐解决长音频转录中的漂移、幻觉及时间戳不准确问题,实现词级时间戳和十二倍批量推理加速,并在长音频转录与词分割基准上达到最先进性能。

摘要:

大规模弱监督语音识别模型(如 Whisper)已在跨领域和跨语言的语音识别中展现出令人印象深刻的结果。然而,它们通过缓冲或滑动窗口方法应用于长音频转录时容易出现漂移、幻觉和重复;并且由于其顺序处理特性,无法进行批量转录。此外,每个话语对应的时间戳往往不准确,词级时间戳也无法开箱即用。为克服这些挑战,我们提出了 WhisperX,一种具有词级时间戳的时间精确语音识别系统,利用语音活动检测和强制音素对齐。通过这种方式,我们在长音频转录和词分割基准上展示了最先进的性能。另外,我们表明使用所提出的 VAD Cut & Merge 策略对音频进行预分割可提高转录质量,并通过批量推理实现十二倍的转录加速。

英文摘要:

Large-scale, weakly-supervised speech recognition models, such as Whisper, have demonstrated impressive results on speech recognition across domains and languages. However, their application to long audio transcription via buffered or sliding window approaches is prone to drifting, hallucination & repetition; and prohibits batched transcription due to their sequential nature. Further, timestamps corresponding each utterance are prone to inaccuracies and word-level timestamps are not available out-of-the-box. To overcome these challenges, we present WhisperX, a time-accurate speech recognition system with word-level timestamps utilising voice activity detection and forced phoneme alignment. In doing so, we demonstrate state-of-the-art performance on long-form transcription and word segmentation benchmarks. Additionally, we show that pre-segmenting audio with our proposed VAD Cut & Merge strategy improves transcription quality and enables a twelve-fold transcription speedup via batched inference.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑