Quick Start of Speech-to-Text

Several shell scripts provided in ./examples/tiny/local will help us to quickly give it a try, for most major modules, including data preparation, model training, case inference, and model evaluation, with a few public datasets (e.g. LibriSpeech, Aishell). Reading these examples will also help you to understand how to make it work with your data.

Some of the scripts in ./examples are not configured with GPUs. If you want to train with 8 GPUs, please modify CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7. If you don't have any GPU available, please set CUDA_VISIBLE_DEVICES= to use CPUs instead. Besides, if an out-of-memory problem occurs, just reduce batch_size to fit.

Let's take a tiny sampled subset of LibriSpeech dataset for instance.

  • Go to the directory

    cd examples/tiny

    Notice that this is only a toy example with a tiny sampled subset of LibriSpeech. If you would like to try with the complete dataset (would take several days for training), please go to examples/librispeech instead.

  • Source env

    source path.sh

    Must do this before you start to do anything. Set MAIN_ROOT as project dir. Using the default deepspeech2 model as MODEL, you can change this in the script.

  • Main entry point

    bash run.sh

    This is just a demo, please make sure every step works well before the next step.

More detailed information is provided in the following sections. Wish you a happy journey with the DeepSpeech on PaddlePaddle ASR engine!

Training a model

The key steps of training for the Mandarin language are the same as that of the English language and we have also provided an example for Mandarin training with Aishell in examples/aishell/local. As mentioned above, please execute sh data.sh, sh train.sh and sh test.sh to do data preparation, training, and testing correspondingly.

Evaluate a Model

To evaluate a model's performance quantitatively, please run:

CUDA_VISIBLE_DEVICES=0 bash local/test.sh

The error rate (default: word error rate; can be set with error_rate_type) will be printed.

We provide two types of CTC decoders: CTC greedy decoder and CTC beam search decoder. The CTC greedy decoder is an implementation of the simple best-path decoding algorithm, selecting at each timestep the most likely token, thus being greedy and locally optimal. The CTC beam search decoder otherwise utilizes a heuristic breadth-first graph search for reaching near-global optimality; it also requires a pre-trained KenLM language model for better scoring and ranking. The decoder type can be set with the argument decoding_method.