This commit is contained in:
Yannik Schmidt
2018-08-15 20:56:35 +02:00
commit 0eb65e4772
8 changed files with 51 additions and 0 deletions

BIN
.plan.txt.swp Normal file

Binary file not shown.

15
plan.txt Normal file
View File

@@ -0,0 +1,15 @@
Smartphone:
- record
- play
- send to server
- recive text form server (?)
Server:
- revice from smartphone
- use google api for recognition
- process/format result (i.e. templating (i.e. templating))
- send resulting text
Computer/Client:
- recivce text
- place it somewhere

BIN
python-server/.main.py.swp Normal file

Binary file not shown.

8
python-server/main.py Executable file
View File

@@ -0,0 +1,8 @@
#!/usr/bin/python3
import s_input
import sys
if __name__ == "__main__":
path = sys.argv[1]
print("AudioFile: {}".format(path))
print(s_input.analyse_file_by_path(path))

6
python-server/s_input.py Normal file
View File

@@ -0,0 +1,6 @@
import speech
def analyse_file_by_path(file_path):
transcript = speech.analyse(file_path)
return transcript

17
python-server/speech.py Normal file
View File

@@ -0,0 +1,17 @@
import speech_recognition as spr
def analyse(file_path):
''' returns the transcripted audio, or None if the analysis fails '''
recognizer = spr.Recognizer()
with sr.AudioFile(file_path) as source:
audio = r.record(source)
try:
string = r.recognize_google(audio)
except spr.UnknownValueError:
log.log("Audio file is broken or not an audio file")
return None
except spr.RequestError as e:
log.log("Could not connect to google API: {}".format(e))
return None
return string

5
setup.sh Normal file
View File

@@ -0,0 +1,5 @@
apt-get install python python-all-dev python-pip build-essential swig git libpulse-dev libasound2-dev
pip3 install wheel
pip3 install google-cloud-speech
pip3 install pocketsphin
pip3 install SpeechRecognition

Binary file not shown.