Speech-to-Text modules

These modules allow you to use various Speech-to-Text services from the Freeswith.
All this modules available from dialplan or scripts through standard Freeswitch API. (see examples below)

For purchase questions, please visit contact page.

mod_sivr_asr
A special module for interaction with sttd.
A pretty flexible solution with a wide range of settings, can be configured for various tasks.
Price: $150 / 150 USDT
mod_google_asr
Allows to use Google speech-to-text service.
Available both commercial and open-source versions.
Price: $150 / 150 USDT
mod_openai_asr
Allows to use OpenAI speech-to-text service.
Available both commercial and open-source versions.
Price: $150 / 150 USDT

Features comparison table of сommercial and free versions.

mod_sivr_asr | mod_google_asr | mod_openai_asr

Feature	Commercial
Supports audio channels	2
FreesSWITCH VAD (voice activity detection)	+
Advanced VAD	+
Speaker identification	+
AGC (auto gain control)	+
JSON response with extra params (audio file, duration, etc)	+
Saving audios (that used for recognition)	+
Source code	-

Feature	Free	Commercial
Supports audio channels	1	2
FreeSWITCH VAD (voice activity detection)	+	+
Advanced VAD	-	+
AGC (auto gain control)	-	+
JSON response with extra params (audio file, duration, etc)	-	+
Saving audios (that used for recognition)	-	+
Flexible HTTP settings (proxy/timeouts/etc...)	+	+
Source code	+	-

Feature	Free	Commercial
Supports audio channels	1	2
FreeSWITCH VAD (voice activity detection)	+	+
Advanced VAD	-	+
AGC (auto gain control)	-	+
JSON response with extra params (audio file, duration, etc)	-	+
Saving audios (that used for recognition)	-	+
Flexible HTTP settings (proxy/timeouts/etc...)	+	+
Source code	+	-

--- Examples ---

Dialplan example:

        
<extension name="asr-test">
 <condition field="destination_number" expression="^(12345)$" >
  <action application="answer" />
  <action application="sleep" data="1000" />
  <action application="play_and_detect_speech" data="/tmp/test2.wav detect:google {lang=en}" />
  <action application="sleep" data="1000" />
  <action application="log" data="INFO SPEECH_RESULT=${detect_speech_result}" />
  <action application="sleep" data="1000" />
  <action application="hangup" />
</condition>
</extension>

Javascript example:

    
session.ttsEngine= 'google'; // requires mod_google_tts
session.asrEngine= 'google';

var txt = session.sayAndDetectSpeech('Hello, how can I help you?', 10);
consoleLog('info', "TEXT: " + txt);

Command line:

    
freeswitch> google_asr_transcript /tmp/test.wav
+OK: How old is the Brooklyn Bridge