ConceptDynamic Time Warping / 动态时间规整
DTW
A dynamic programming-based similarity measure for time series that finds an optimal warping path to align sequences with differing lengths or tempos. DTW is widely used in speech recognition, gesture matching, and related domains, but suffers from high computational cost and non-differentiability.
Timeline (last 90 days)
Oct 15
In audio regression testing, because TTS models are non-deterministic, the same text synthesized twice produces different byte streams but sounds consistent, so byte-level comparison cannot be used; instead, acoustic fingerprints (such as MFCC, F0 trajectory, duration) should be extracted and compared with golden audio using cosine similarity or DTW distance.
Unverified50%