experiments/STT
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Update evaluate_tflite with wav_filename

Browse Source
This commit is contained in:
Alexandre Lissy 2019-10-31 15:41:11 +01:00
parent f3694efbca
commit f3240bffbc
1 changed files with 16 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
evaluate_tflite.py
View File

@ -36,18 +36,21 @@ def tflite_worker(model, alphabet, lm, trie, queue_in, queue_out, gpu_mask):
ds.enableDecoderWithLM(lm, trie, LM_ALPHA, LM_BETA) ds.enableDecoderWithLM(lm, trie, LM_ALPHA, LM_BETA)
while True: while True:
msg = queue_in.get() try:
msg = queue_in.get()
filename = msg['filename'] filename = msg['filename']
wavname = os.path.splitext(os.path.basename(filename))[0] wavname = os.path.splitext(os.path.basename(filename))[0]
fin = wave.open(filename, 'rb') fin = wave.open(filename, 'rb')
fs = fin.getframerate() audio = np.frombuffer(fin.readframes(fin.getnframes()), np.int16)
audio = np.frombuffer(fin.readframes(fin.getnframes()), np.int16) fin.close()
fin.close()
decoded = ds.stt(audio) decoded = ds.stt(audio)
queue_out.put({'wav': wavname, 'prediction': decoded, 'ground_truth': msg['transcript']})
except FileNotFoundError as ex:
print('FileNotFoundError: ', ex)
queue_out.put({'wav': wavname, 'prediction': decoded, 'ground_truth': msg['transcript']})
print(queue_out.qsize(), end='\r') # Update the current progress print(queue_out.qsize(), end='\r') # Update the current progress
queue_in.task_done() queue_in.task_done()
@ -85,6 +88,7 @@ def main():
ground_truths = [] ground_truths = []
predictions = [] predictions = []
losses = [] losses = []
wav_filenames = []
with open(args.csv, 'r') as csvfile: with open(args.csv, 'r') as csvfile:
csvreader = csv.DictReader(csvfile) csvreader = csv.DictReader(csvfile)
@ -92,6 +96,8 @@ def main():
for row in csvreader: for row in csvreader:
count += 1 count += 1
work_todo.put({'filename': row['wav_filename'], 'transcript': row['transcript']}) work_todo.put({'filename': row['wav_filename'], 'transcript': row['transcript']})
wav_filenames.extend(row['wav_filename'])
print('Totally %d wav entries found in csv\n' % count) print('Totally %d wav entries found in csv\n' % count)
work_todo.join() work_todo.join()
print('\nTotally %d wav file transcripted' % work_done.qsize()) print('\nTotally %d wav file transcripted' % work_done.qsize())
@ -103,7 +109,7 @@ def main():
predictions.append(msg['prediction']) predictions.append(msg['prediction'])
wavlist.append(msg['wav']) wavlist.append(msg['wav'])
wer, cer, _ = calculate_report(ground_truths, predictions, losses) wer, cer, samples = calculate_report(wav_filenames, ground_truths, predictions, losses)
mean_loss = np.mean(losses) mean_loss = np.mean(losses)
print('Test - WER: %f, CER: %f, loss: %f' % print('Test - WER: %f, CER: %f, loss: %f' %