Vosk speech models bundled in build/models/ (vosk-model-small-ru-0.22,
vosk-model-small-en-us-0.15) are provided by Alpha Cephei Inc. under the
Apache License 2.0 (same as Vosk itself — see vosk.txt for the full text).

Copyright 2020 Alpha Cephei Inc. <https://alphacephei.com/vosk/models>

--- vosk-model-small-en-us-0.15/README ---
US English model for mobile Vosk applications

Copyright 2020 Alpha Cephei Inc

Accuracy: 10.38 (tedlium test) 9.85 (librispeech test-clean)
Speed: 0.11xRT (desktop)
Latency: 0.15s (right context)



--- vosk-model-small-ru-0.22/README ---
Small Russian model for Vosk (Android, RPi, other small devices)

%WER 22.71 [ 9092 / 40042, 1124 ins, 1536 del, 6432 sub ] exp/chain_a/tdnn/decode_test_audiobooks_look_fast/wer_10_0.0
%WER 11.79 [ 5940 / 50394, 894 ins, 832 del, 4214 sub ] exp/chain_a/tdnn/decode_test_golos_crowd_look_fast/wer_11_0.0
%WER 21.34 [ 1789 / 8382, 173 ins, 440 del, 1176 sub ] exp/chain_a/tdnn/decode_test_golos_farfield_look_fast/wer_10_0.0
%WER 29.89 [ 5579 / 18666, 476 ins, 1550 del, 3553 sub ] exp/chain_a/tdnn/decode_test_sova_devices_look_fast/wer_10_0.0
%WER 31.97 [ 13588 / 42496, 1013 ins, 3640 del, 8935 sub ] exp/chain_a/tdnn/decode_test_youtube_look_fast/wer_9_0.0

