CODEWITHSUNDEEP
python-3.xspeech-recognitionpocketsphinx
pbb.business
pbb.business

Reputation: 21

Where to place trained Speech model

I have trained pocketsphinx using sphinxtrain. The problem I'm facing is that I don't know how to use the trained model in my code.

My first thought was to just replace the current model in the pocketsphinx library or to include the trained model somehow.

I have searched a lot but most of what I found was based around using tensorflow for training or Googles recognition software, but nothing about how to use a trained model.

Here is a basic example of how the code works:

import speech_recognition as sr


r = sr.Recognizer()
with sr.Microphone() as source:
    r.adjust_for_ambient_noise(source)
    audio = r.listen(source)



output = r.recognize_sphinx(audio)
print(output)

Upvotes: 1

Views: 171

Answers (1)

pbb.business
pbb.business

Reputation: 21

I solved the problem by using pocketsphinx's LiveSpeech() 

import os
from pocketsphinx import LiveSpeech, get_model_path
model_path = get_model_path()

speech = LiveSpeech(
    verbose=False,
    sampling_rate=16000,
    buffer_size=2048,
    no_search=False,
    full_utt=False,
    hmm=os.path.join(model_path, 'en-us'),
    lm=os.path.join(model_path, 'en-us.lm.bin'),
    dic=os.path.join(model_path, 'cmudict-en-us.dict')
)

for phrase in speech:
    output = phrase.hypothesis()

    if output == 'hello':
        print("recognized")
        print(output)
    else:
        print("not recognized")
        print(output)

In this example the output should look something like this for the if statement 

recognized
hello

and like this for the else statement

not recognized
hi

Upvotes: 1

Related Questions