hub / github.com/tensorflow/lingvo / ExtractLogMelFeatures

Function ExtractLogMelFeatures

lingvo/tools/audio_lib.py:71–111 · view source on GitHub ↗

Create Log-Mel Filterbank Features from raw bytes. Args: wav_bytes_t: Tensor representing raw wav file as a string of bytes. It is currently assumed that the wav file is encoded at 16KHz (see DecodeWav, below). Returns: A Tensor representing three stacked log-Mel filterbank

(wav_bytes_t)

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers

nothing calls this directly

Calls 3

DecodeWavFunction · 0.85

_CreateAsrFrontendFunction · 0.85

FPropDefaultThetaMethod · 0.45

Tested by

no test coverage detected

Used in the wild real call sites across dependent graphs

searching dependent graphs…