I am working on
- Training or just running inference - training/fine-tuning
- DeepSpeech branch/version - 0.7.0
- OS Platform and Distribution (e.g., Linux Ubuntu 18.04) - Linux Ubuntu 16.04
- Python version -3.6.5
- TensorFlow version - 1.15.2
- CUDA/cuDNN version - CUDA 10.0/cuDNN 7.6.5
Was interested in knowing if the released 0.7.0 checkpoint can detect “um”, “uh” given that it was trained on Fisher and SWB which are conversational datasets. Also in the case of fine tuning should my training data and transcription have enough examples ( with the correct um and uh transcriptions ) to be able to detected these as well.