github.com/mozilla/DSAlign issues | Ecosyste.ms: OpenCollective

Phoneme-level alignment

DanielSWolf opened this issue over 3 years ago

TaskCluster Download Issue

kaleko opened this issue over 3 years ago

Quick and dirty #36 fix

comodoro opened this pull request about 4 years ago

Tensorflow warning

ishan-modi opened this issue about 4 years ago

Could not generate example data from bin/gettestdata.sh

OswaldoBornemann opened this issue about 4 years ago

Is there any way to speed the alignment process?

SenYan1999 opened this issue about 4 years ago

Alphabet is not defined despite being loaded.

ArEnSc opened this issue over 4 years ago

Updated documentation and minor tool fixes

tilmankamp opened this pull request over 4 years ago

./bin/lm-dependencies.sh is failing because some files are missing

shivamiitgoa opened this issue over 4 years ago

Part of aligned text gets shifted to the next segment

BoneGoat opened this issue over 4 years ago

Support for DeepSpeech 0.7.x (WIP)

BoneGoat opened this pull request over 4 years ago

Update DSAlign to 0.7+

madziszyn opened this issue over 4 years ago

Dataset/catalog access

nicolaspanel opened this issue over 4 years ago

max is missing

Jendker opened this pull request over 4 years ago

Duplicating audio/text pairs

carlfm01 opened this issue over 4 years ago

Do not load model if it's already loaded

gibiansky opened this pull request over 4 years ago

Large catalog spawns many processes which won't die

BoneGoat opened this issue almost 5 years ago

inconsistencies

Tortoise17 opened this issue almost 5 years ago

handling long sentences

Tortoise17 opened this issue almost 5 years ago

Skip inputs with empty clean transcripts

lunixbochs opened this pull request almost 5 years ago

seems to hang if transcript contains no alphabet characters

lunixbochs opened this issue almost 5 years ago

alphabet issues

Tortoise17 opened this issue almost 5 years ago

Excluded {} empty transcripts error

lunixbochs opened this issue almost 5 years ago

ERROR: 4-gram discount out of range for adjusted count 3: -0.9085922

krupalraj opened this issue almost 5 years ago

Fix KenLM model generation on smaller texts

lunixbochs opened this pull request almost 5 years ago

Fix align.sh example in README

lunixbochs opened this pull request almost 5 years ago

Fix for Mac OSX

BoneGoat opened this issue about 5 years ago

Feature/deepspeech 0.6.0

BoneGoat opened this pull request about 5 years ago

Adapt for DeepSpeech 0.6.0

BoneGoat opened this issue about 5 years ago

Use other STT systems ?

noe opened this issue about 5 years ago

raising StopIteration causes a RunTime error on Python 3.8

tensorfoo opened this issue about 5 years ago

MemoryError when calling align on a large audio file (~1gb mp3)

tensorfoo opened this issue about 5 years ago

aligned output json doesn't have unicode text

tensorfoo opened this issue about 5 years ago

Unable to generate document specific language model

bwang482 opened this issue about 5 years ago

Doesn't use GPU

bwang482 opened this issue about 5 years ago

trying to run the example in the readme.md

tensorfoo opened this issue about 5 years ago

word level alignment

Tortoise17 opened this issue about 5 years ago

assigning speaker

Tortoise17 opened this issue about 5 years ago

StopIteration error while aligning

PaulLerner opened this issue about 5 years ago

INFO:root:Fragment 11: character error rate (CER) too high

Aradhya0510 opened this issue over 5 years ago

where is buildkenlm.sh?

msubhash opened this issue over 5 years ago

Is the transcript.txt in data/test1 correct ?

Kimonode opened this issue over 5 years ago