SeanNaren / deepspeech.pytorch

Speech Recognition using DeepSpeech2.

SeanNaren/deepspeech.pytorch Issues

it did not generates the models/deepspeech.pth file
Updated 2 months ago2
SequenceWise Operation
Updated 4 months ago
Support for newest version of pytorch-lightning right now (2.1)
Updated 6 months ago
How can I train or test without spectrogram?
Closed 7 months ago1
Got too high WER and CER on AN4
Closed 7 months ago1
What should we do if we have segments for dataset?
Updated 10 months ago1
TypeError("__init__() got an unexpected keyword argument 'num_processes'")
Updated 10 months ago3
partition_activations produces no activation memory improvement with zero3
Updated 10 months ago1
Training
Updated 10 months ago
deepspeech2 need how much Flops?
Closed 2 years ago1
How to support microphone vad by deepspeech.pytorch?
Closed 2 years ago1
noise inject
Closed 3 years ago3
error in training
Closed 2 years ago1
loss (Loss 0.00000) for all along the training
Closed 2 years ago4
Does batch size change while traininig with elastic agent?
Closed 2 years ago1
weights_summary and weights_path on lightning_config not working.
Closed 2 years ago1
About omegaconf.errors.ValidationError: Unexpected object type: tuple at train_config.py
Closed 2 years ago2
omegaconf.errors.ValidationError: Unexpected object type: tuple
Closed 2 years ago1
Question about data shuffling
Closed 2 years ago3
Should we remove Hydra in favour of LightningCLI?
Closed 2 years ago3
English speech dataset
Closed 2 years ago1
/bin/sh: 1: sox: not found
Closed 2 years ago2
cant train!
Closed 2 years ago1
could you please provide a docker image with an established environment?
Closed 2 years ago1
Key 'period' not in 'ModelCheckpointConf'
Closed 2 years ago2
Transcribe
Closed 2 years ago1
about pytorch_lightning
Closed 2 years ago1
decoding issue
Closed 2 years ago1
Platform in dependent
Closed 2 years ago1
integrate with Lightning ecosystem CI
Closed 2 years ago
Hydra error : config_path is not specified in @hydra.main()
Closed 3 years ago4
Not able to load the saved Checkpoints
Closed 3 years ago4
Trainning process has been stopped after 4th epoch
Closed 3 years ago1
Deployment on Hardware?
Closed 3 years ago1
FineTune with new label.json
Closed 3 years ago1
v3.0 gets higher CER and WER than v2.1 while training on custom dataset
Closed 3 years ago5
Training is interrupted without error with 3 gpus
Closed 3 years ago1
beam search decoder in LM
Closed 3 years ago1
TypeError: __init__() got an unexpected keyword argument 'prefix'
Closed 3 years ago14
The feature extraction technique (or the input to the neural network model) used in this implementation?
Closed 3 years ago1
How to train custom dataset on pretrained model with v3.0?
Closed 3 years ago1
Facing type error while giving the command for training
Closed 3 years ago8
beam search
Closed 3 years ago1
RuntimeError: solve_cpu: For batch 0: U(4,4) is zero, singular U.
Closed 3 years ago1
Error occur after running this commend python "train.py data.train_path=data/train_manifest.json data.val_path=data/val_manifest.json"
Closed 3 years ago2
Very high WER on custom testing-data
Closed 3 years ago3
how long does it take to achieve the accurate
Closed 3 years ago3
inference predicts meaningless letters rather than sentence
Closed 3 years ago2
Has anybody converted this to coreml without using ONNX?
Closed 3 years ago1
Cannot run Inference Server
Closed 3 years ago