acoustic:
    left_pad: 0.3  # in sec
    right_pad: 0.3  # in sec
    chunk_size: 0.3  # in sec
    state_size: 219729
    frame_size: 0.03  # in sec
    bias: 0.39  # in sec
    sample_rate: 8000
    labels: ["а", "б", "в", "г", "д", "е", "ё", "ж", "з", "и", "й", "к", "л", "м", "н", "о", "п",
             "р", "с", "т", "у", "ф", "х", "ц", "ч", "ш", "щ", "ъ", "ы", "ь", "э", "ю", "я", " "]

splitter:
    labels: ${acoustic.labels}
    speech_offset: 0.1  # in sec
    silence_duration_threshold: 0.6  # in sec
    acoustic_silence_prob_threshold: 0.9  # in sec
    min_acoustic_silence_frames: 3
    phrase_duration_threshold: 60.  # in sec; split very long phrase by force 

decoder:
    decoding_strategy: "beam_search"   # greedy_search | beam_search
    beam_search_params:
        beam_width: 200
        alpha: 0.4
        beta: 0.9
        unigrams: null