distillpub / post--ctc
Sequence Modelling with CTC
☆45Updated last year
Related projects: ⓘ
- Auto Segmentation Criterion (ASG) implemented in pytorch☆50Updated 2 years ago
- PyTorch bindings for Warp-CTC☆42Updated 4 years ago
- Tensor2tensor experiment with SpecAugment☆47Updated 5 years ago
- Losses and decoders for end-to-end ASR and OCR☆33Updated 3 years ago
- A simple implementation of the paper https://arxiv.org/pdf/1910.00716v1.pdf☆31Updated 2 years ago
- PyTorch Implementations for End-to-End Automatic Speech Recognition☆126Updated 5 years ago
- Example implementation of Monotonic Chunkwise Attention.☆49Updated 6 years ago
- Speech recognition☆8Updated 4 years ago
- ☆12Updated 7 years ago
- Attention, I'm Trying to Speak: End-to-end speech synthesis (CS224n '18)☆52Updated 5 years ago
- PyTorch CTC Decoder bindings☆14Updated 6 years ago
- Tensorflow with KenLM integrated for beam search scoring☆34Updated 7 years ago
- Source code for 'Transfer Learning for Speech Recognition on a Budget' published at ACL 2017☆46Updated 7 years ago
- Sound augmentation using Large-scale audio dataset (Audioset)☆44Updated 3 years ago
- ☆13Updated this week
- Speech2vec pre-trained word vectors☆77Updated 6 years ago
- ☆74Updated 2 years ago
- Efficient DataLoader for PyTorch and Keras for loading datasets from web servers and object stores.☆29Updated 4 years ago
- ctc_beamsearch☆18Updated 7 years ago
- Pytorch implementation of sparse_image_warp and an example of GoogleBrain's SpecAugment is given: A Simple Data Augmentation Method for A…☆22Updated 5 years ago
- An implementation of Tacotron and Tacotron2☆81Updated 3 years ago
- Python API for reading and querying ARPA formatted language models.☆33Updated 10 years ago
- An Attention Based Open-Source End to End Speech Synthesis Framework, No CNN, No RNN, No MFCC!!!☆85Updated 3 years ago
- Python implementation of CTC beam search decoder + agnostic LM scorer☆19Updated 3 years ago
- compare three CTC decoder, that is greedy decoder, beam decoder and prefix beam decoder☆19Updated 6 years ago
- The implementation of 'Watch, Listen, Attend and Spell’ (WLAS) network that learns to transcribe videos of mouth motion to character on p…☆11Updated 6 years ago
- The Additive Margin SincNet (AM-SincNet) is a new approach for speaker recognition problems which is based in the neural network architec…☆43Updated 11 months ago
- ESPnet-TTS Audio Sample HP☆21Updated 4 years ago
- ☆43Updated this week
- Compute useful transcriptions metrics (CER, WER, SER, ...)☆26Updated 9 years ago