espnet-model-zoo 0.0.0a9

ESPnet Model Zoo

ESPnet Model Zoo

Utilities managing the pretrained models created by ESPnet.

• Zenodo community: https://zenodo.org/communities/espnet/
• Registered models: table.csv

Install

pip install torch
pip install espnet_model_zoo

Obtain a model

from espnet_model_zoo.downloader import ModelDownloader
d = ModelDownloader("~/.cache/espnet")  # Specify cachedir
d = ModelDownloader()  # <module_dir> is used as cachedir by default

To obtain a model, you need to give a model in the form of <user_name/model_name>

>>> d.download_and_unpack("kamo-naoyuki/mini_an4_asr_train_raw_bpe_valid.acc.best")
{"asr_train_config": <config path>, "asr_model_file": <model path>, ...}

Note that if the model already exists, you can skip downloading and unpacking.

You can also get a model with certain conditions.

d.download_and_unpack(task="asr", corpus="wsj")

If multiple models are found with the condition, the last model is selected. You can also specify the condition using "version" option.

d.download_and_unpack(task="asr", corpus="wsj", version=-1)  # Get the last model
d.download_and_unpack(task="asr", corpus="wsj", version=-2)  # Get previous model

You can also obtain it from the URL directly.

d.download_and_unpack("https://zenodo.org/record/...")

Query model names

You can view the model names from our Zenodo community, https://zenodo.org/communities/espnet/, or using query(). All information are written in table.csv.

d.query("name")

You can also show them with specifying certain conditions.

d.query("name", task="asr")

Command line tools

• espnet_model_zoo_query

  # Query model name
  espnet_model_zoo_query --condition task=asr --condition corpus=wsj 
  # Query the other key
  espnet_model_zoo_query --key url --condition task=asr --condition corpus=wsj 
  

• espnet_model_zoo_download

  espnet_model_zoo_download <model_name>  # Print the path of the downloaded file
  espnet_model_zoo_download --unpack true <model_name>   # Print the path of unpacked files
  

• espnet_model_zoo_upload

  export ACCESS_TOKEN=<access_token>
  espnet_zenodo_upload \
      --file <packed_model> \
      --title <title> \
      --description <description> \
      --creator_name <your-git-account>
  

Use a pretrained model for inference

ASR

import soundfile
from espnet_model_zoo.downloader import ModelDownloader
from espnet2.bin.asr_inference import Speech2Text
d = ModelDownloader()
speech2text = Speech2Text(**d.download_and_unpack("user_name/model_name"))

speech, rate = soundfile.read("speech.wav")
speech2text(speech)
[(text, token, token_int, hypothesis object), ...]

TTS

import soundfile
from espnet_model_zoo.downloader import ModelDownloader
from espnet2.bin.tts_inference import Text2Speech
d = ModelDownloader()
text2speech = Text2Speech(**d.download_and_unpack("user_name/model_name"))

retval = text2speech("foobar")
speech = retval[0]
soundfile.write("out.wav", speech.numpy(), text2speech.fs, "PCM_16")

Register your model

1. Upload your model to Zenodo

   You need to signup to Zenodo and create an access token to upload models. You can upload your own model by using espnet_model_zoo_upload command freely, but we normally upload a model using recipes.

2. Create a Pull Request to modify table.csv

   You need to append your record at the last line.

3. (Administrator does) Increment the third version number of setup.py, e.g. 0.0.3 -> 0.0.4

4. (Administrator does) Release new version

Update your model

If your model has some troubles, please modify the record at Zenodo directly or reupload a corrected file using espnet_zenodo_upload as another record.