Skip to content
Script to interact with tensorflow speech on Ubuntu/Raspberry pi3
Python
Branch: master
Clone or download
Pull request Compare This branch is 25 commits ahead of jeysonmc:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.gitignore
README.md
_config.yml
audio_intensity.py
audio_intensity_trigger.py
audio_trigger_save_wav.py
mod_label_wav.py
pyAudio_wav.py
stt_google.py
wav_trigger_inference.py

README.md

Forked from jeysonmc/python-google-speech-scripts. Edits are for removing google speech recognition and divert it to local tensorflow speech recognition

Speech detection with Tensorflow 1.4 on Raspberry Pi 3 – Part 2: Live audio inferencing using PyAudio

http://www.kiranjose.in/blogs/speech-detection-with-tensorflow-1-4-on-raspberry-pi-3-part-2-live-audio-inferencing-using-pyaudio/

Usage:

cd /tensorflow/examples/speech_commands

touch file.wav ; to create a dummy file for the first pass

python3 wav_trigger_inference.py --graph=./my_frozen_graph.pb --labels=./conv_labels.txt --wav=file.wav

You can’t perform that action at this time.