Not logged in.  Login/Logout/Register | List snippets | | Create snippet | Upload image | Upload data

35
LINES

< > BotCompany Repo | #1026985 // recognize.py (vosk)

Document

#!/usr/bin/python3

from vosk import Model, KaldiRecognizer
import sys
import json
import os

if not os.path.exists(sys.argv[1]):
    print ("Please download the model from https://github.com/alphacep/kaldi-android-demo/releases and unpack as 'model' in the current folder.")
    exit (1)

model = Model(sys.argv[1])

# Large vocabulary free form recognition
rec = KaldiRecognizer(model, 16000)

# You can also specify the possible word list
#rec = KaldiRecognizer(model, 16000, "zero oh one two three four five six seven eight nine")

wf = open(sys.argv[2], "rb")
wf.read(44) # skip header

while True:
    data = wf.read(2000)
    if len(data) == 0:
        break
    if rec.AcceptWaveform(data):
        res = json.loads(rec.Result())
        print (res)
    else:
        res = json.loads(rec.PartialResult())
        print (res)

res = json.loads(rec.FinalResult())
print (res)

download  show line numbers   

Travelled to 8 computer(s): bhatertpkbcr, mqqgnosmbjvj, pyentgdyhuwx, pzhvpgtvlbxg, snaazhdonpnp, tvejysmllsmz, vouqrxazstgt, xrpafgyirdlv

No comments. add comment

Snippet ID: #1026985
Snippet name: recognize.py (vosk)
Eternal ID of this version: #1026985/1
Text MD5: 5a5a8aa6739017f843f5c63a56669d22
Author: stefan
Category: javax / audio recognition
Type: Document
Public (visible to everyone): Yes
Archived (hidden from active list): No
Created/modified: 2020-02-06 02:28:33
Source code size: 916 bytes / 35 lines
Pitched / IR pitched: No / No
Views / Downloads: 328 / 185
Referenced in: #1026981 - vosk_recognize16KWav