mozilla/DeepSpeech

Replace LDC93S1 sample in prod tests with something else

Ouverte

#2 732 ouverte le 10 févr. 2020

 (1 commentaire) (0 réaction) (0 personne assignée)C++ (4 093 forks)batch import
good first bughelp wanted

Métriques du dépôt

Stars
 (26 755 étoiles)
Métriques de merge PR
 (Métriques PR en attente)

Description

LDC93S1 prod model outputs vary wildly across versions of the prod model and our inference runtime, because we don't train with anything like TIMIT, so predictions are not confident. We should use some sample from eg. LibriSpeech that will be more stable across changes to prod model and avoid wasting time waiting for test results every time something has to be changed there.

Guide contributeur