37 lines
1.0 KiB
Plaintext
37 lines
1.0 KiB
Plaintext
|
#!/usr/bin/env python3
|
||
|
|
|
||
|
|
from vosk import Model, KaldiRecognizer, SetLogLevel
|
||
|
|
import sys
|
||
|
|
import json
|
||
|
|
import os
|
||
|
|
import wave
|
||
|
|
import subprocess
|
||
|
|
|
||
|
|
SetLogLevel(0)
|
||
|
|
|
||
|
|
modelBase = '/home/janek/data/projects/_forks/vosk-api/python/example/models'
|
||
|
|
modelPath = os.path.join(modelBase, sys.argv[2] if len(sys.argv) > 2 else 'daanzu')
|
||
|
|
if not os.path.exists(modelPath):
|
||
|
|
print (f"Please download a model from https://alphacephei.com/vosk/models and unpack it to {modelPath}.", file=sys.stderr)
|
||
|
|
exit (1)
|
||
|
|
|
||
|
|
sample_rate=16000
|
||
|
|
model = Model(modelPath)
|
||
|
|
rec = KaldiRecognizer(model, sample_rate)
|
||
|
|
|
||
|
|
process = subprocess.Popen(['ffmpeg', '-loglevel', 'quiet', '-i',
|
||
|
|
sys.argv[1],
|
||
|
|
'-ar', str(sample_rate) , '-ac', '1', '-f', 's16le', '-'],
|
||
|
|
stdout=subprocess.PIPE)
|
||
|
|
|
||
|
|
while True:
|
||
|
|
data = process.stdout.read(4000)
|
||
|
|
if len(data) == 0:
|
||
|
|
break
|
||
|
|
if rec.AcceptWaveform(data):
|
||
|
|
res = json.loads(rec.Result())
|
||
|
|
print (res['text'])
|
||
|
|
|
||
|
|
res = json.loads(rec.FinalResult())
|
||
|
|
print (res['text'])
|