diff options
Diffstat (limited to 'README.md')
| -rw-r--r-- | README.md | 7 |
1 files changed, 4 insertions, 3 deletions
@@ -1,5 +1,5 @@ # Automotive Grade Linux (AGL) Voice Agent Service -A gRPC-based voice agent service designed for Automotive Grade Linux (AGL). This service leverages GStreamer, Vosk, Snips, and RASA to seamlessly process user voice commands. It converts spoken words into text, extracts intents from these commands, and performs actions through the Kuksa interface. +A gRPC-based voice agent service designed for Automotive Grade Linux (AGL). This service leverages GStreamer, Vosk, Whisper, Snips, and RASA to seamlessly process user voice commands. It converts spoken words into text, extracts intents from these commands, and performs actions through the Kuksa interface. ## Table of Contents - [Features](#features) @@ -62,14 +62,15 @@ Replace `SERVER_IP` with IP address of the running Voice Agent server, and `SERV To issue a voice command, use the following command: ```bash -voiceagent-service run-client --server_address SERVER_IP --server_port SERVER_PORT --action ExecuteVoiceCommand --mode manual --nlu NLU_ENGINE +voiceagent-service run-client --server_address SERVER_IP --server_port SERVER_PORT --action ExecuteVoiceCommand --mode manual --nlu NLU_ENGINE --stt-frameword STT_FRAMEWORK ``` -Replace `NLU_ENGINE` with the preferred NLU engine ("snips" or "rasa"), `SERVER_IP` with IP address of the running Voice Agent server, and `SERVER_PORT` with the port of the running Voice Agent server. You can also pass a custom value to flag `--recording-time` if you want to change the default recording time from 5 seconds to any other value. +Replace `NLU_ENGINE` with the preferred NLU engine ("snips" or "rasa"), `SERVER_IP` with IP address of the running Voice Agent server, and `SERVER_PORT` with the port of the running Voice Agent server. You can also pass a custom value to flag `--recording-time` if you want to change the default recording time from 5 seconds to any other value, you can also pass --stt-frameword to specify the STT framework to be used, supported frameworks are "vosk" and "whisper" and the default is "vosk". ## Configuration Configuration options for the AGL Voice Agent Service can be found in the default `config.ini` file. You can customize various settings, including the AI models, audio directories, and Kuksa integration. **Important:** while manually making changes to the config file make sure you add trailing slash to all the directory paths, ie. the paths to directories should always end with a `/`. ## Maintainers +- **Anuj Solanki** <anuj603362@gmail.com> - **Malik Talha** <talhamalik727x@gmail.com> ## License |