#!/usr/bin/python3 # -*- coding: utf-8 -*- from enum import Enum from functools import lru_cache import logging import os import platform from pathlib import Path import huggingface_hub import sherpa import sherpa_onnx main_logger = logging.getLogger("main") class EnumDecodingMethod(Enum): greedy_search = "greedy_search" modified_beam_search = "modified_beam_search" model_map = { "Chinese": [ { "repo_id": "csukuangfj/wenet-chinese-model", "nn_model_file": "final.zip", "nn_model_file_sub_folder": ".", "tokens_file": "units.txt", "tokens_file_sub_folder": ".", "normalize_samples": False, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "csukuangfj/sherpa-onnx-paraformer-zh-2024-03-09", "nn_model_file": "model.int8.onnx", "nn_model_file_sub_folder": ".", "tokens_file": "tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_offline_recognizer_from_paraformer", }, { "repo_id": "csukuangfj/sherpa-onnx-paraformer-zh-small-2024-03-09", "nn_model_file": "model.int8.onnx", "nn_model_file_sub_folder": ".", "tokens_file": "tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_offline_recognizer_from_paraformer", }, { "repo_id": "luomingshuang/icefall_asr_wenetspeech_pruned_transducer_stateless2", "nn_model_file": "cpu_jit_epoch_10_avg_2_torch_1.7.1.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_char", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "zrjin/sherpa-onnx-zipformer-multi-zh-hans-2023-9-2", "encoder_model_file": "encoder-epoch-20-avg-1.onnx", "encoder_model_file_sub_folder": ".", "decoder_model_file": "decoder-epoch-20-avg-1.onnx", "decoder_model_file_sub_folder": ".", "joiner_model_file": "joiner-epoch-20-avg-1.onnx", "joiner_model_file_sub_folder": ".", "tokens_file": "tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_offline_recognizer_from_transducer", }, { "repo_id": "zrjin/icefall-asr-aishell-zipformer-large-2023-10-24", "encoder_model_file": "encoder-epoch-56-avg-23.onnx", "encoder_model_file_sub_folder": "exp", "decoder_model_file": "decoder-epoch-56-avg-23.onnx", "decoder_model_file_sub_folder": "exp", "joiner_model_file": "joiner-epoch-56-avg-23.onnx", "joiner_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_char", "loader": "load_sherpa_onnx_offline_recognizer_from_transducer", }, { "repo_id": "zrjin/icefall-asr-aishell-zipformer-small-2023-10-24", "encoder_model_file": "encoder-epoch-55-avg-21.onnx", "encoder_model_file_sub_folder": "exp", "decoder_model_file": "decoder-epoch-55-avg-21.onnx", "decoder_model_file_sub_folder": "exp", "joiner_model_file": "joiner-epoch-55-avg-21.onnx", "joiner_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_char", "loader": "load_sherpa_onnx_offline_recognizer_from_transducer", }, { "repo_id": "zrjin/icefall-asr-aishell-zipformer-2023-10-24", "encoder_model_file": "encoder-epoch-55-avg-17.onnx", "encoder_model_file_sub_folder": "exp", "decoder_model_file": "decoder-epoch-55-avg-17.onnx", "decoder_model_file_sub_folder": "exp", "joiner_model_file": "joiner-epoch-55-avg-17.onnx", "joiner_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_char", "loader": "load_sherpa_onnx_offline_recognizer_from_transducer", }, { "repo_id": "desh2608/icefall-asr-alimeeting-pruned-transducer-stateless7", "nn_model_file": "cpu_jit.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_char", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "yuekai/icefall-asr-aishell2-pruned-transducer-stateless5-A-2022-07-12", "nn_model_file": "cpu_jit.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_char", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "yuekai/icefall-asr-aishell2-pruned-transducer-stateless5-B-2022-07-12", "nn_model_file": "cpu_jit.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_char", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "luomingshuang/icefall_asr_aidatatang-200zh_pruned_transducer_stateless2", "nn_model_file": "cpu_jit_torch.1.7.1.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_char", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "luomingshuang/icefall_asr_alimeeting_pruned_transducer_stateless2", "nn_model_file": "cpu_jit_torch_1.7.1.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_char", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, ], "English": [ { "repo_id": "csukuangfj/sherpa-onnx-whisper-tiny.en", "encoder_model_file": "tiny.en-encoder.int8.onnx", "encoder_model_file_sub_folder": ".", "decoder_model_file": "tiny.en-decoder.int8.onnx", "decoder_model_file_sub_folder": ".", "tokens_file": "tiny.en-tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_offline_recognizer_from_whisper", }, { "repo_id": "csukuangfj/sherpa-onnx-whisper-base.en", "encoder_model_file": "base.en-encoder.int8.onnx", "encoder_model_file_sub_folder": ".", "decoder_model_file": "base.en-decoder.int8.onnx", "decoder_model_file_sub_folder": ".", "tokens_file": "base.en-tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_offline_recognizer_from_whisper", }, { "repo_id": "csukuangfj/sherpa-onnx-whisper-small.en", "encoder_model_file": "small.en-encoder.int8.onnx", "encoder_model_file_sub_folder": ".", "decoder_model_file": "small.en-decoder.int8.onnx", "decoder_model_file_sub_folder": ".", "tokens_file": "small.en-tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_offline_recognizer_from_whisper", }, { "repo_id": "csukuangfj/sherpa-onnx-paraformer-en-2024-03-09", "nn_model_file": "model.int8.onnx", "nn_model_file_sub_folder": ".", "tokens_file": "tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_offline_recognizer_from_paraformer", }, { "repo_id": "yfyeung/icefall-asr-gigaspeech-zipformer-2023-10-17", "encoder_model_file": "encoder-epoch-30-avg-9.onnx", "encoder_model_file_sub_folder": "exp", "decoder_model_file": "decoder-epoch-30-avg-9.onnx", "decoder_model_file_sub_folder": "exp", "joiner_model_file": "joiner-epoch-30-avg-9.onnx", "joiner_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "loader": "load_sherpa_onnx_offline_recognizer_from_transducer", }, { "repo_id": "wgb14/icefall-asr-gigaspeech-pruned-transducer-stateless2", "nn_model_file": "cpu_jit-iter-3488000-avg-20.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "./giga-tokens.txt", "tokens_file_sub_folder": ".", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "yfyeung/icefall-asr-multidataset-pruned_transducer_stateless7-2023-05-04", "nn_model_file": "cpu_jit-epoch-30-avg-4.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "yfyeung/icefall-asr-finetune-mux-pruned_transducer_stateless7-2023-05-19", "nn_model_file": "cpu_jit-epoch-20-avg-5.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "WeijiZhuang/icefall-asr-librispeech-pruned-transducer-stateless8-2022-12-02", "nn_model_file": "cpu_jit-torch-1.10.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "csukuangfj/icefall-asr-librispeech-pruned-transducer-stateless8-2022-11-14", "nn_model_file": "cpu_jit.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "csukuangfj/icefall-asr-librispeech-pruned-transducer-stateless7-2022-11-11", "nn_model_file": "cpu_jit-torch-1.10.0.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "csukuangfj/icefall-asr-librispeech-pruned-transducer-stateless3-2022-05-13", "nn_model_file": "cpu_jit.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "yujinqiu/sherpa-onnx-paraformer-en-2023-10-24", "nn_model_file": "model.int8.onnx", "nn_model_file_sub_folder": ".", "tokens_file": "new_tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_offline_recognizer_from_paraformer", }, { "repo_id": "Zengwei/icefall-asr-librispeech-zipformer-large-2023-05-16", "nn_model_file": "jit_script.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "Zengwei/icefall-asr-librispeech-zipformer-2023-05-15", "nn_model_file": "jit_script.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "Zengwei/icefall-asr-librispeech-zipformer-small-2023-05-16", "nn_model_file": "jit_script.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "videodanchik/icefall-asr-tedlium3-conformer-ctc2", "nn_model_file": "cpu_jit.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "pkufool/icefall_asr_librispeech_conformer_ctc", "nn_model_file": "cpu_jit.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "WayneWiser/icefall-asr-librispeech-conformer-ctc2-jit-bpe-500-2022-07-21", "nn_model_file": "cpu_jit.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "csukuangfj/wenet-english-model", "nn_model_file": "final.zip", "nn_model_file_sub_folder": ".", "tokens_file": "units.txt", "tokens_file_sub_folder": ".", "normalize_samples": False, "loader": "load_sherpa_offline_recognizer", }, ], "Chinese+English": [ { "repo_id": "csukuangfj/sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20", "encoder_model_file": "encoder-epoch-99-avg-1.onnx", "encoder_model_file_sub_folder": ".", "decoder_model_file": "decoder-epoch-99-avg-1.onnx", "decoder_model_file_sub_folder": ".", "joiner_model_file": "joiner-epoch-99-avg-1.onnx", "joiner_model_file_sub_folder": ".", "tokens_file": "tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_online_recognizer_from_transducer", }, { "repo_id": "csukuangfj/sherpa-onnx-paraformer-zh-2023-03-28", "nn_model_file": "model.int8.onnx", "nn_model_file_sub_folder": ".", "tokens_file": "tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_offline_recognizer_from_paraformer", }, { "repo_id": "ptrnull/icefall-asr-conv-emformer-transducer-stateless2-zh", "nn_model_file": "cpu_jit-epoch-11-avg-1.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_char_bpe", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "luomingshuang/icefall_asr_tal-csasr_pruned_transducer_stateless5", "nn_model_file": "cpu_jit.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_char", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, ], "Chinese+English+Cantonese": [ { "repo_id": "csukuangfj/sherpa-onnx-paraformer-trilingual-zh-cantonese-en", "nn_model_file": "model.int8.onnx", "nn_model_file_sub_folder": ".", "tokens_file": "tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_offline_recognizer_from_paraformer", }, { "repo_id": "csukuangfj/sherpa-onnx-streaming-paraformer-trilingual-zh-cantonese-en", "encoder_model_file": "encoder.int8.onnx", "encoder_model_file_sub_folder": ".", "decoder_model_file": "decoder.int8.onnx", "decoder_model_file_sub_folder": ".", "tokens_file": "tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_online_recognizer_from_paraformer", }, ], "Cantonese": [ { "repo_id": "zrjin/icefall-asr-mdcc-zipformer-2024-03-11", "encoder_model_file": "encoder-epoch-45-avg-35.int8.onnx", "encoder_model_file_sub_folder": "exp", "decoder_model_file": "decoder-epoch-45-avg-35.onnx", "decoder_model_file_sub_folder": "exp", "joiner_model_file": "joiner-epoch-45-avg-35.int8.onnx", "joiner_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_char", "loader": "load_sherpa_onnx_offline_recognizer_from_transducer", }, ], # "Japanese": [ # { # "repo_id": "TeoWenShen/icefall-asr-csj-pruned-transducer-stateless7-streaming-230208-fluent", # "encoder_model_file": "encoder_jit_trace.pt", # "encoder_model_file_sub_folder": "exp_fluent", # "decoder_model_file": "decoder_jit_trace.pt", # "decoder_model_file_sub_folder": "exp_fluent", # "joiner_model_file": "joiner_jit_trace.pt", # "joiner_model_file_sub_folder": "exp_fluent", # "tokens_file": "tokens.txt", # "tokens_file_sub_folder": "data/lang_char", # "normalize_samples": True, # "loader": "load_sherpa_online_recognizer", # }, # { # "repo_id": "TeoWenShen/icefall-asr-csj-pruned-transducer-stateless7-streaming-230208-disfluent", # "encoder_model_file": "encoder_jit_trace.pt", # "encoder_model_file_sub_folder": "exp_disfluent", # "decoder_model_file": "decoder_jit_trace.pt", # "decoder_model_file_sub_folder": "exp_disfluent", # "joiner_model_file": "joiner_jit_trace.pt", # "joiner_model_file_sub_folder": "exp_disfluent", # "tokens_file": "tokens.txt", # "tokens_file_sub_folder": "data/lang_char", # "normalize_samples": True, # "loader": "load_sherpa_online_recognizer", # }, # ], "German": [ { "repo_id": "csukuangfj/wav2vec2.0-torchaudio", "nn_model_file": "voxpopuli_asr_base_10k_de.pt", "nn_model_file_sub_folder": ".", "tokens_file": "tokens-de.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_offline_recognizer_without_feat_config", }, ], "French": [ { "repo_id": "shaojieli/sherpa-onnx-streaming-zipformer-fr-2023-04-14", "encoder_model_file": "encoder-epoch-29-avg-9-with-averaged-model.onnx", "encoder_model_file_sub_folder": ".", "decoder_model_file": "decoder-epoch-29-avg-9-with-averaged-model.onnx", "decoder_model_file_sub_folder": ".", "joiner_model_file": "joiner-epoch-29-avg-9-with-averaged-model.onnx", "joiner_model_file_sub_folder": ".", "tokens_file": "tokens.txt", "tokens_file_sub_folder": ".", "loader": "load_sherpa_onnx_online_recognizer_from_transducer", }, ], "Russian": [ { "repo_id": "alphacep/vosk-model-ru", "encoder_model_file": "encoder.onnx", "encoder_model_file_sub_folder": "am-onnx", "decoder_model_file": "decoder.onnx", "decoder_model_file_sub_folder": "am-onnx", "joiner_model_file": "joiner.onnx", "joiner_model_file_sub_folder": "am-onnx", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "lang", "loader": "load_sherpa_onnx_offline_recognizer_from_transducer", }, { "repo_id": "alphacep/vosk-model-small-ru", "encoder_model_file": "encoder.onnx", "encoder_model_file_sub_folder": "am", "decoder_model_file": "decoder.onnx", "decoder_model_file_sub_folder": "am", "joiner_model_file": "joiner.onnx", "joiner_model_file_sub_folder": "am", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "lang", "loader": "load_sherpa_onnx_offline_recognizer_from_transducer", }, ], "Arabic": [ { "repo_id": "AmirHussein/icefall-asr-mgb2-conformer_ctc-2022-27-06", "nn_model_file": "cpu_jit.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_5000", "loader": "load_sherpa_offline_recognizer_without_feat_config", }, ], "Tibetan": [ { "repo_id": "syzym/icefall-asr-xbmu-amdo31-pruned-transducer-stateless7-2022-12-02", "nn_model_file": "cpu_jit.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, { "repo_id": "syzym/icefall-asr-xbmu-amdo31-pruned-transducer-stateless5-2022-11-29", "nn_model_file": "cpu_jit-epoch-28-avg-23-torch-1.10.0.pt", "nn_model_file_sub_folder": "exp", "tokens_file": "tokens.txt", "tokens_file_sub_folder": "data/lang_bpe_500", "normalize_samples": True, "loader": "load_sherpa_offline_recognizer", }, ], } def download_model(local_model_dir: str, **kwargs, ): repo_id = kwargs["repo_id"] if "nn_model_file" in kwargs.keys(): main_logger.info("download nn_model_file. filename: {}, subfolder: {}".format(kwargs["nn_model_file"], kwargs["nn_model_file_sub_folder"])) _ = huggingface_hub.hf_hub_download( repo_id=repo_id, filename=kwargs["nn_model_file"], subfolder=kwargs["nn_model_file_sub_folder"], local_dir=local_model_dir, ) if "encoder_model_file" in kwargs.keys(): main_logger.info("download encoder_model_file. filename: {}, subfolder: {}".format(kwargs["encoder_model_file"], kwargs["encoder_model_file_sub_folder"])) _ = huggingface_hub.hf_hub_download( repo_id=repo_id, filename=kwargs["encoder_model_file"], subfolder=kwargs["encoder_model_file_sub_folder"], local_dir=local_model_dir, ) if "decoder_model_file" in kwargs.keys(): main_logger.info("download decoder_model_file. filename: {}, subfolder: {}".format(kwargs["decoder_model_file"], kwargs["decoder_model_file_sub_folder"])) _ = huggingface_hub.hf_hub_download( repo_id=repo_id, filename=kwargs["decoder_model_file"], subfolder=kwargs["decoder_model_file_sub_folder"], local_dir=local_model_dir, ) if "joiner_model_file" in kwargs.keys(): main_logger.info("download joiner_model_file. filename: {}, subfolder: {}".format(kwargs["joiner_model_file"], kwargs["joiner_model_file_sub_folder"])) _ = huggingface_hub.hf_hub_download( repo_id=repo_id, filename=kwargs["joiner_model_file"], subfolder=kwargs["joiner_model_file_sub_folder"], local_dir=local_model_dir, ) if "tokens_file" in kwargs.keys(): main_logger.info("download tokens_file. filename: {}, subfolder: {}".format(kwargs["tokens_file"], kwargs["tokens_file_sub_folder"])) tokens_file = kwargs["tokens_file"] if not tokens_file.startswith("./"): _ = huggingface_hub.hf_hub_download( repo_id=repo_id, filename=kwargs["tokens_file"], subfolder=kwargs["tokens_file_sub_folder"], local_dir=local_model_dir, ) def load_sherpa_offline_recognizer(nn_model_file: str, tokens_file: str, sample_rate: int = 16000, num_active_paths: int = 2, decoding_method: str = "greedy_search", num_mel_bins: int = 80, frame_dither: int = 0, normalize_samples: bool = False, ): feat_config = sherpa.FeatureConfig(normalize_samples=normalize_samples) feat_config.fbank_opts.frame_opts.samp_freq = sample_rate feat_config.fbank_opts.mel_opts.num_bins = num_mel_bins feat_config.fbank_opts.frame_opts.dither = frame_dither if not os.path.exists(nn_model_file): raise AssertionError("nn_model_file not found. nn_model_file: {}".format(nn_model_file)) config = sherpa.OfflineRecognizerConfig( nn_model=nn_model_file, tokens=tokens_file, use_gpu=False, feat_config=feat_config, decoding_method=decoding_method, num_active_paths=num_active_paths, ) recognizer = sherpa.OfflineRecognizer(config) return recognizer def load_sherpa_offline_recognizer_without_feat_config(nn_model_file: str, tokens_file: str, num_active_paths: int = 2, decoding_method: str = "greedy_search", ): config = sherpa.OfflineRecognizerConfig( nn_model=nn_model_file, tokens=tokens_file, use_gpu=False, decoding_method=decoding_method, num_active_paths=num_active_paths, ) recognizer = sherpa.OfflineRecognizer(config) return recognizer def load_sherpa_onnx_offline_recognizer_from_paraformer(nn_model_file: str, tokens_file: str, sample_rate: int = 16000, decoding_method: str = "greedy_search", feature_dim: int = 80, num_threads: int = 2, ): recognizer = sherpa_onnx.OfflineRecognizer.from_paraformer( paraformer=nn_model_file, tokens=tokens_file, num_threads=num_threads, sample_rate=sample_rate, feature_dim=feature_dim, decoding_method=decoding_method, debug=False, ) return recognizer def load_sherpa_onnx_offline_recognizer_from_transducer(encoder_model_file: str, decoder_model_file: str, joiner_model_file: str, tokens_file: str, sample_rate: int = 16000, decoding_method: str = "greedy_search", feature_dim: int = 80, num_threads: int = 2, num_active_paths: int = 2, ): recognizer = sherpa_onnx.OfflineRecognizer.from_transducer( encoder=encoder_model_file, decoder=decoder_model_file, joiner=joiner_model_file, tokens=tokens_file, num_threads=num_threads, sample_rate=sample_rate, feature_dim=feature_dim, decoding_method=decoding_method, max_active_paths=num_active_paths, ) return recognizer def load_sherpa_onnx_offline_recognizer_from_whisper(encoder_model_file: str, decoder_model_file: str, tokens_file: str, num_threads: int = 2, ): recognizer = sherpa_onnx.OfflineRecognizer.from_whisper( encoder=encoder_model_file, decoder=decoder_model_file, tokens=tokens_file, num_threads=num_threads, ) return recognizer def load_sherpa_online_recognizer(nn_model_file: str, encoder_model_file: str, decoder_model_file: str, joiner_model_file: str, tokens_file: str, sample_rate: int = 16000, num_active_paths: int = 2, decoding_method: str = "greedy_search", num_mel_bins: int = 80, frame_dither: int = 0, normalize_samples: bool = False, ): feat_config = sherpa.FeatureConfig(normalize_samples=normalize_samples) feat_config.fbank_opts.frame_opts.samp_freq = sample_rate feat_config.fbank_opts.mel_opts.num_bins = num_mel_bins feat_config.fbank_opts.frame_opts.dither = frame_dither if not os.path.exists(nn_model_file): raise AssertionError("nn_model_file not found. nn_model_file: {}".format(nn_model_file)) config = sherpa.OfflineRecognizerConfig( nn_model=nn_model_file, encoder_model=encoder_model_file, decoder_model=decoder_model_file, joiner_model=joiner_model_file, tokens=tokens_file, use_gpu=False, feat_config=feat_config, decoding_method=decoding_method, num_active_paths=num_active_paths, chunk_size=32, ) recognizer = sherpa.OnlineRecognizer(config) return recognizer def load_sherpa_onnx_online_recognizer_from_transducer(encoder_model_file: str, decoder_model_file: str, joiner_model_file: str, tokens_file: str, sample_rate: int = 16000, decoding_method: str = "greedy_search", feature_dim: int = 80, num_threads: int = 2, num_active_paths: int = 2, ): recognizer = sherpa_onnx.OnlineRecognizer.from_transducer( encoder=encoder_model_file, decoder=decoder_model_file, joiner=joiner_model_file, tokens=tokens_file, num_threads=num_threads, sample_rate=sample_rate, feature_dim=feature_dim, decoding_method=decoding_method, max_active_paths=num_active_paths, ) return recognizer def load_sherpa_onnx_online_recognizer_from_paraformer(encoder_model_file: str, decoder_model_file: str, tokens_file: str, sample_rate: int = 16000, decoding_method: str = "greedy_search", feature_dim: int = 80, num_threads: int = 2, ): recognizer = sherpa_onnx.OnlineRecognizer.from_paraformer( encoder=encoder_model_file, decoder=decoder_model_file, tokens=tokens_file, num_threads=num_threads, sample_rate=sample_rate, feature_dim=feature_dim, decoding_method=decoding_method, ) return recognizer @lru_cache(maxsize=15) def load_recognizer(local_model_dir: Path, decoding_method: str = "greedy_search", num_active_paths: int = 4, **kwargs, ): if not local_model_dir.exists(): download_model( local_model_dir=local_model_dir.as_posix(), **kwargs, ) loader = kwargs["loader"] kwargs_ = dict() if "nn_model_file" in kwargs.keys(): nn_model_file = (local_model_dir / kwargs["nn_model_file_sub_folder"] / kwargs["nn_model_file"]).as_posix() kwargs_["nn_model_file"] = nn_model_file if "encoder_model_file" in kwargs.keys(): encoder_model_file = (local_model_dir / kwargs["encoder_model_file_sub_folder"] / kwargs["encoder_model_file"]).as_posix() kwargs_["encoder_model_file"] = encoder_model_file if "decoder_model_file" in kwargs.keys(): decoder_model_file = (local_model_dir / kwargs["decoder_model_file_sub_folder"] / kwargs["decoder_model_file"]).as_posix() kwargs_["decoder_model_file"] = decoder_model_file if "joiner_model_file" in kwargs.keys(): joiner_model_file = (local_model_dir / kwargs["joiner_model_file_sub_folder"] / kwargs["joiner_model_file"]).as_posix() kwargs_["joiner_model_file"] = joiner_model_file if "tokens_file" in kwargs.keys(): tokens_file: str = kwargs["tokens_file"] if not tokens_file.startswith("./"): tokens_file = (local_model_dir / kwargs["tokens_file_sub_folder"] / kwargs["tokens_file"]).as_posix() kwargs_["tokens_file"] = tokens_file if "normalize_samples" in kwargs.keys(): kwargs_["normalize_samples"] = kwargs["normalize_samples"] if loader == "load_sherpa_offline_recognizer": recognizer = load_sherpa_offline_recognizer( decoding_method=decoding_method, num_active_paths=num_active_paths, **kwargs_ ) elif loader == "load_sherpa_offline_recognizer_without_feat_config": recognizer = load_sherpa_offline_recognizer_without_feat_config( decoding_method=decoding_method, **kwargs_ ) elif loader == "load_sherpa_onnx_offline_recognizer_from_paraformer": recognizer = load_sherpa_onnx_offline_recognizer_from_paraformer( decoding_method=decoding_method, **kwargs_ ) elif loader == "load_sherpa_onnx_offline_recognizer_from_transducer": recognizer = load_sherpa_onnx_offline_recognizer_from_transducer( decoding_method=decoding_method, **kwargs_ ) elif loader == "load_sherpa_onnx_offline_recognizer_from_whisper": recognizer = load_sherpa_onnx_offline_recognizer_from_whisper( **kwargs_ ) elif loader == "load_sherpa_online_recognizer": recognizer = load_sherpa_online_recognizer( decoding_method=decoding_method, num_active_paths=num_active_paths, **kwargs_ ) elif loader == "load_sherpa_onnx_online_recognizer_from_transducer": recognizer = load_sherpa_onnx_online_recognizer_from_transducer( **kwargs_ ) elif loader == "load_sherpa_onnx_online_recognizer_from_paraformer": recognizer = load_sherpa_onnx_online_recognizer_from_paraformer( **kwargs_ ) else: raise NotImplementedError("loader not support: {}".format(loader)) return recognizer @lru_cache(maxsize=15) def load_punctuation_model(local_model_dir: Path, repo_id: str, nn_model_file: str, nn_model_file_sub_folder: str, ): if not local_model_dir.exists(): download_model( local_model_dir=local_model_dir.as_posix(), repo_id=repo_id, nn_model_file=nn_model_file, nn_model_file_sub_folder=nn_model_file_sub_folder, ) nn_model_file = (local_model_dir / nn_model_file_sub_folder / nn_model_file).as_posix() config = sherpa_onnx.OfflinePunctuationConfig( model=sherpa_onnx.OfflinePunctuationModelConfig( ct_transformer=nn_model_file ), ) punctuation_model = sherpa_onnx.OfflinePunctuation(config) return punctuation_model if __name__ == "__main__": pass