ESPnet: end-to-end speech processing toolkit

Docs | Example | Docker | Notebook | Tutorial (2019)

ESPnet is an end-to-end speech processing toolkit, mainly focuses on end-to-end speech recognition and end-to-end text-to-speech. ESPnet uses chainer and pytorch as a main deep learning engine, and also follows Kaldi style data processing, feature extraction/format, and recipes to provide a complete setup for speech recognition and other speech processing experiments.

Key Features

Kaldi style complete recipe

• Support numbers of ASR recipes (WSJ, Switchboard, CHiME-4/5, Librispeech, TED, CSJ, AMI, HKUST, Voxforge, REVERB, etc.)
• Support numbers of TTS recipes with a similar manner to the ASR recipe (LJSpeech, LibriTTS, M-AILABS, etc.)
• Support numbers of ST recipes (Fisher-CallHome Spanish, Libri-trans, IWSLT'18, How2, Must-C, Mboshi-French, etc.)
• Support numbers of MT recipes (IWSLT'16, the above ST recipes etc.)
• Support speech separation and recognition recipe (WSJ-2mix)

ASR: Automatic Speech Recognition

• State-of-the-art performance in several ASR benchmarks (comparable/superior to hybrid DNN/HMM and CTC)
• Hybrid CTC/attention based end-to-end ASR
  • Fast/accurate training with CTC/attention multitask training
  • CTC/attention joint decoding to boost monotonic alignment decoding
  • Encoder: VGG-like CNN + BiRNN (LSTM/GRU), sub-sampling BiRNN (LSTM/GRU) or Transformer
• Attention: Dot product, location-aware attention, variants of multihead
• Incorporate RNNLM/LSTMLM/TransformerLM trained only with text data
• Batch GPU decoding
• Transducer based end-to-end ASR
  • Available: RNN-Transducer, Transformer-Transducer, Transformer/RNN-Transducer
  • Support attention extension and VGG-Transformer (encoder)

TTS: Text-to-speech

• Tacotron2 based end-to-end TTS
• Transformer based end-to-end TTS
• Feed-forward Transformer (a.k.a. FastSpeech) based end-to-end TTS (new!)

ST: Speech Translation & MT: Machine Translation

• State-of-the-art performance in several ST benchmarks (comparable/superior to cascaded ASR and MT)
• Transformer based end-to-end ST (new!)
• Transformer based end-to-end MT (new!)

DNN Framework

• Flexible network architecture thanks to chainer and pytorch
• Flexible front-end processing thanks to kaldiio and HDF5 support
• Tensorboard based monitoring

Installation

See https://espnet.github.io/espnet/installation.html

Usage

See https://espnet.github.io/espnet/tutorial.html

Docker Container

go to docker/ and follow instructions.

Contribution

Any contributions to ESPNet are welcome and feel free to ask any questions or requests to issues. If you are the first commiter, please follow the contribution guide.

Results and demo

You can find useful tutorials and demos in Interspeech 2019 Tutorial

ASR results

We list the character error rate (CER) and word error rate (WER) of major ASR tasks.

Task	CER (%)	WER (%)	Pretrained model
Aishell dev	6.0	N/A	link
Aishell test	6.7	N/A	same as above
Common Voice dev	1.7	2.2	link
Common Voice test	1.8	2.3	same as above
CSJ eval1	5.7	N/A	link
CSJ eval2	3.8	N/A	same as above
CSJ eval3	4.2	N/A	same as above
HKUST dev	23.5	N/A	link
Librispeech dev_clean	N/A	2.1	link
Librispeech dev_other	N/A	5.3	same as above
Librispeech test_clean	N/A	2.5	same as above
Librispeech test_other	N/A	5.5	same as above
TEDLIUM2 dev	N/A	9.3	link
TEDLIUM2 test	N/A	8.1	same as above
TEDLIUM3 dev	N/A	9.7	link
TEDLIUM3 test	N/A	8.0	same as above
WSJ dev93	3.2	7.0	N/A
WSJ eval92	2.1	4.7	N/A

Note that the performance of the CSJ, HKUST, and Librispeech tasks was significantly improved by using the wide network (#units = 1024) and large subword units if necessary reported by RWTH.

If you want to check the results of the other recipes, please check egs/<name_of_recipe>/asr1/RESULTS.md.

ASR demo

You can recognize speech in a WAV file using pretrained models. Go to a recipe directory and run utils/recog_wav.sh as follows:

cd egs/tedlium2/asr1
../../../utils/recog_wav.sh --models tedlium2.transformer.v1 example.wav

where example.wav is a WAV file to be recognized. The sampling rate must be consistent with that of data used in training.

Available pretrained models in the demo script are listed as below.

Model	Notes
tedlium2.rnn.v1	Streaming decoding based on CTC-based VAD
tedlium2.rnn.v2	Streaming decoding based on CTC-based VAD (batch decoding)
tedlium2.transformer.v1	Joint-CTC attention Transformer trained on Tedlium 2
tedlium3.transformer.v1	Joint-CTC attention Transformer trained on Tedlium 3
librispeech.transformer.v1	Joint-CTC attention Transformer trained on Librispeech
commonvoice.transformer.v1	Joint-CTC attention Transformer trained on CommonVoice
csj.transformer.v1	Joint-CTC attention Transformer trained on CSJ

ST results

We list 4-gram BLEU of major ST tasks.

end-to-end system

Task	BLEU	Pretrained model
Fisher-CallHome Spanish fisher_test (Es->En)	48.39	link
Fisher-CallHome Spanish callhome_evltest (Es->En)	18.67	link
Libri-trans test (En->Fr)	16.70	link
How2 dev5 (En->Pt)	45.68	link
Must-C tst-COMMON (En->De)	22.91	link
Mboshi-French dev (Fr->Mboshi)	6.18	N/A

cascaded system

Task	BLEU	Pretrained model
Fisher-CallHome Spanish fisher_test (Es->En)	42.16	N/A
Fisher-CallHome Spanish callhome_evltest (Es->En)	19.82	N/A
Libri-trans test (En->Fr)	16.96	N/A
How2 dev5 (En->Pt)	44.90	N/A
Must-C tst-COMMON (En->De)	23.65	N/A

If you want to check the results of the other recipes, please check egs/<name_of_recipe>/st1/RESULTS.md.

ST demo

(New!) We made a new real-time E2E-ST + TTS demonstration in Google Colab. Please access the notebook from the following button and enjoy the real-time speech-to-speech translation!

---

You can translate speech in a WAV file using pretrained models. Go to a recipe directory and run utils/translate_wav.sh as follows:

cd egs/fisher_callhome_spanish/st1/
wget -O - https://github.com/espnet/espnet/files/4100928/test.wav.tar.gz | tar zxvf - ../../../utils/translate_wav.sh --models fisher_callhome_spanish.transformer.v1.es-en test.wav

where test.wav is a WAV file to be translated. The sampling rate must be consistent with that of data used in training.

Available pretrained models in the demo script are listed as below.

Model	Notes
fisher_callhome_spanish.transformer.v1	Transformer-ST trained on Fisher-CallHome Spanish Es->En

MT results

Task	BLEU	Pretrained model
Fisher-CallHome Spanish fisher_test (Es->En)	61.45	link
Fisher-CallHome Spanish callhome_evltest (Es->En)	29.86	link
Libri-trans test (En->Fr)	18.09	link
How2 dev5 (En->Pt)	58.61	link
Must-C tst-COMMON (En->De)	27.63	link
IWSLT'14 test2014 (En->De)	24.70	link
IWSLT'14 test2014 (De->En)	29.22	link
IWSLT'16 test2014 (En->De)	24.05	link
IWSLT'16 test2014 (De->En)	29.13	link

TTS results

You can listen to our samples in demo HP espnet-tts-sample. Here we list some notable ones:

• Single English speaker Tacotron2
• Single Japanese speaker Tacotron2
• Single other language speaker Tacotron2
• Multi English speaker Tacotron2
• Single English speaker Transformer
• Single English speaker FastSpeech
• Multi English speaker Transformer
• Single Italian speaker FastSpeech
• Single Mandarin speaker Transformer
• Single Mandarin speaker FastSpeech
• Multi Japanese speaker Transformer
• Single English speaker models with Parallel WaveGAN
• Single English speaker knowledge distillation-based FastSpeech (New!)

You can download all of the pretrained models and generated samples:

• All of the pretrained E2E-TTS models
• All of the generated samples

Note that in the generated samples we use three vocoders: Griffin-Lim (GL), WaveNet vocoder (WaveNet), Parallel WaveGAN (ParallelWaveGAN), and MelGAN (MelGAN). The neural vocoders are based on following repositories.

• kan-bayashi/ParallelWaveGAN: Parallel WaveGAN / MelGAN
• r9y9/wavenet_vocoder: 16 bit mixture of Logistics WaveNet vocoder
• kan-bayashi/PytorchWaveNetVocoder: 8 bit Softmax WaveNet Vocoder with the noise shaping

If you want to build your own neural vocoder, please check the above repositories.

Here we list all of the pretrained neural vocoders. Please download and enjoy the generation of high quality speech!

Model link	Lang	Fs [Hz]	Mel range [Hz]	FFT / Shift / Win [pt]	Model type
ljspeech.wavenet.softmax.ns.v1	EN	22.05k	None	1024 / 256 / None	Softmax WaveNet
ljspeech.wavenet.mol.v1	EN	22.05k	None	1024 / 256 / None	MoL WaveNet
ljspeech.parallel_wavegan.v1	EN	22.05k	None	1024 / 256 / None	Parallel WaveGAN
ljspeech.wavenet.mol.v2	EN	22.05k	80-7600	1024 / 256 / None	MoL WaveNet
ljspeech.parallel_wavegan.v2	EN	22.05k	80-7600	1024 / 256 / None	Parallel WaveGAN
ljspeech.melgan.v1 (EXPERIMENTAL)	EN	22.05k	80-7600	1024 / 256 / None	MelGAN
libritts.wavenet.mol.v1	EN	24k	None	1024 / 256 / None	MoL WaveNet
jsut.wavenet.mol.v1	JP	24k	80-7600	2048 / 300 / 1200	MoL WaveNet
jsut.parallel_wavegan.v1	JP	24k	80-7600	2048 / 300 / 1200	Parallel WaveGAN
csmsc.wavenet.mol.v1	ZH	24k	80-7600	2048 / 300 / 1200	MoL WaveNet
csmsc.parallel_wavegan.v1	ZH	24k	80-7600	2048 / 300 / 1200	Parallel WaveGAN

If you want to use the above pretrained vocoders, please exactly match the feature setting with them.

TTS demo

(New!) We made a new real-time E2E-TTS demonstration in Google Colab. Please access the notebook from the following button and enjoy the real-time synthesis!

---

You can synthesize speech in a TXT file using pretrained models. Go to a recipe directory and run utils/synth_wav.sh as follows:

cd egs/ljspeech/tts1
echo "THIS IS A DEMONSTRATION OF TEXT TO SPEECH." > example.txt
../../../utils/synth_wav.sh example.txt

You can change the pretrained model as follows:

../../../utils/synth_wav.sh --models ljspeech.fastspeech.v1 example.txt

Waveform synthesis is performed with Griffin-Lim algorithm and neural vocoders (WaveNet and ParallelWaveGAN). You can change the pretrained vocoder model as follows:

../../../utils/synth_wav.sh --vocoder_models ljspeech.wavenet.mol.v1 example.txt

Note that WaveNet vocoder provides very high quality speech but it takes time to generate.

Available pretrained models in the demo script are listed as follows:

Model link	Lang	Fs [Hz]	Mel range [Hz]	FFT / Shift / Win [pt]	Input	R	Model type
ljspeech.tacotron2.v1	EN	22.05k	None	1024 / 256 / None	char	2	Tacotron 2
ljspeech.tacotron2.v2	EN	22.05k	None	1024 / 256 / None	char	1	Tacotron 2 + forward attention
ljspeech.tacotron2.v3	EN	22.05k	None	1024 / 256 / None	char	1	Tacotron 2 + guided attention loss
ljspeech.transformer.v1	EN	22.05k	None	1024 / 256 / None	char	1	Deep Transformer
ljspeech.transformer.v2	EN	22.05k	None	1024 / 256 / None	char	3	Shallow Transformer
ljspeech.transformer.v3	EN	22.05k	None	1024 / 256 / None	phn	1	Deep Transformer
ljspeech.fastspeech.v1	EN	22.05k	None	1024 / 256 / None	char	1	FF-Transformer
ljspeech.fastspeech.v2	EN	22.05k	None	1024 / 256 / None	char	1	FF-Transformer + CNN in FFT block
ljspeech.fastspeech.v3	EN	22.05k	None	1024 / 256 / None	phn	1	FF-Transformer + CNN in FFT block + postnet
libritts.tacotron2.v1	EN	24k	80-7600	1024 / 256 / None	char	2	Multi-speaker Tacotron 2
libritts.transformer.v1	EN	24k	80-7600	1024 / 256 / None	char	2	Multi-speaker Transformer
jsut.tacotron2	JP	24k	80-7600	2048 / 300 / 1200	phn	2	Tacotron 2
jsut.transformer	JP	24k	80-7600	2048 / 300 / 1200	phn	3	Shallow Transformer
csmsc.transformer.v1	ZH	24k	80-7600	2048 / 300 / 1200	pinyin	1	Deep Transformer
csmsc.fastspeech.v3	ZH	24k	80-7600	2048 / 300 / 1200	pinyin	1	FF-Transformer + CNN in FFT block + postnet

Available pretrained vocoder models in the demo script are listed as follows:

Model link	Lang	Fs [Hz]	Mel range [Hz]	FFT / Shift / Win [pt]	Model type
ljspeech.wavenet.softmax.ns.v1	EN	22.05k	None	1024 / 256 / None	Softmax WaveNet
ljspeech.wavenet.mol.v1	EN	22.05k	None	1024 / 256 / None	MoL WaveNet
ljspeech.parallel_wavegan.v1	EN	22.05k	None	1024 / 256 / None	Parallel WaveGAN
libritts.wavenet.mol.v1	EN	24k	None	1024 / 256 / None	MoL WaveNet
jsut.wavenet.mol.v1	JP	24k	80-7600	2048 / 300 / 1200	MoL WaveNet
jsut.parallel_wavegan.v1	JP	24k	80-7600	2048 / 300 / 1200	Parallel WaveGAN
csmsc.wavenet.mol.v1	ZH	24k	80-7600	2048 / 300 / 1200	MoL WaveNet
csmsc.parallel_wavegan.v1	ZH	24k	80-7600	2048 / 300 / 1200	Parallel WaveGAN

References

[1] Shinji Watanabe, Takaaki Hori, Shigeki Karita, Tomoki Hayashi, Jiro Nishitoba, Yuya Unno, Nelson Enrique Yalta Soplin, Jahn Heymann, Matthew Wiesner, Nanxin Chen, Adithya Renduchintala, and Tsubasa Ochiai, "ESPnet: End-to-End Speech Processing Toolkit," Proc. Interspeech'18, pp. 2207-2211 (2018)

[2] Suyoun Kim, Takaaki Hori, and Shinji Watanabe, "Joint CTC-attention based end-to-end speech recognition using multi-task learning," Proc. ICASSP'17, pp. 4835--4839 (2017)

[3] Shinji Watanabe, Takaaki Hori, Suyoun Kim, John R. Hershey and Tomoki Hayashi, "Hybrid CTC/Attention Architecture for End-to-End Speech Recognition," IEEE Journal of Selected Topics in Signal Processing, vol. 11, no. 8, pp. 1240-1253, Dec. 2017

Citations

@inproceedings{watanabe2018espnet,
  author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson {Enrique Yalta Soplin} and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai},
  title={ESPnet: End-to-End Speech Processing Toolkit},
  year=2018,
  booktitle={Interspeech},
  pages={2207--2211},
  doi={10.21437/Interspeech.2018-1456},
  url={http://dx.doi.org/10.21437/Interspeech.2018-1456}
}
@misc{hayashi2019espnettts,
    title={ESPnet-TTS: Unified, Reproducible, and Integratable Open Source End-to-End Text-to-Speech Toolkit},
    author={Tomoki Hayashi and Ryuichi Yamamoto and Katsuki Inoue and Takenori Yoshimura and Shinji Watanabe and Tomoki Toda and Kazuya Takeda and Yu Zhang and Xu Tan},
    year={2019},
    eprint={1910.10909},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}