chrsmj / vosk-asterisk

Speech Recognition in Asterisk with Vosk Server

Vosk speech recognition modules for Asterisk

This is an asterisk module for Vosk API server:

https://github.com/alphacep/vosk-server

It is tested with latest asterisk git master, but should equally work with other branches (13,16,17).

Installation

1. Make sure you have latest asterisk update

git clone https://github.com/asterisk/asterisk
....

2. First build the modules

./bootstrap
./configure --with-asterisk=<path_to_asterisk_source> --prefix=<path_to_install>
make
make install

for example:

./bootstrap
./configure --with-asterisk=/usr --prefix=/usr
make
make install

3. Edit modules.conf to load modules

load = res_speech.so
load = res_http_websocket.so
load = res_speech_vosk.so

4. Edit dialplan in extensions.conf:

[internal]
exten = 1,1,Answer
same = n,Wait(1)
same = n,SpeechCreate
same = n,SpeechBackground(hello)
same = n,Verbose(0,Result was ${SPEECH_TEXT(0)})

5. Run Vosk server with the Docker

docker run -d -p 2700:2700 alphacep/kaldi-en:latest

6. Dial extension and check the result