/utbots_voice

text-to-speech & speech-to-text

Primary LanguageC

utbots_voice: ROS packages of text-to-speech (TTS) and speech-to-text (STT)

  • For TTS:
    • We have a ROS interface of Coqui TTS (Mozilla TTS successor, bleeding edge TTS that supports a variety of models).
    • We have a ROS interface of Mimic3 (works well in english and other languages but no portuguese at this moment).
    • You probably don't need two TTS programs. For example, Mimic3 will be very good if only english is needed.
  • For STT:
    • We have a ROS interface of whisper.cpp (high-performance implementation of OpenAI Whisper).

Setup

  • Clone repository and compile workspace

    # Clone repository
cd ~/catkin_ws/src
git clone --recurse-submodules https://github.com/UtBotsAtHome-UTFPR/utbots_voice.git

# Compile workspace
cd ~/catkin_ws
catkin_make
source devel/setup.bash