speechkit 1.2.2

It's lib for using speechkit api by yandex.

speechkit lib python

It's lib for using speechkit api by yandex.

This lib supports short and long audio recognition of speechkit

Install

pip install git+https://github.com/TikhonP/yandex-speechkit-lib-python.git

or

git clone https://github.com/TikhonP/yandex-speechkit-lib-python.git
sudo python setup.py install

How to use

Import:

from speechkit import speechkit

For short audio

Init:

spchkt = speechkit.RecognizeShortAudio(token)

Use token for getting iam

Recognize:

text = spchkt.recognize(filename, folder_id)

Please write folder_id

For long audio

For long audio you need to upload file to the yandex object storage. You can do it with objectStorage class:

objstor = speechkit.ObjectStorage(aws_access_key_id, aws_secret_access_key)

Speechkit works with ogg opus file. This function recoding audio with ffmpeg. It works only on unix system and you need to install ffmpeg.

out = speechkit.recode(infilename, outfilename)

It will save opus file to the same directory.

There is also supporting removing files by command rm

out = speechkit.removefile(inputfile)

They return '0', if process was done successfully

For synthesis

Init:

synthesizeAudio = speechkit.SynthesizeAudio(apiKey, catalogId

Use token and catalogID for getting iam

Synthesize and save to file

synthesizeAudio.synthesize(text, filepath, lpcm=False, voice='alena')

Synthesize and get bufferio data

data = synthesizeAudio.synthesize_stream(text, lpcm=False, voice='alena')