ModelScope/modelscope/models/audio/asr/wenet_automatic_speech_recognition.py

# Copyright (c) Alibaba, Inc. and its affiliates.

import os
from typing import Any, Dict

import json
import wenetruntime as wenet

from modelscope.metainfo import Models
from modelscope.models.base import Model
from modelscope.models.builder import MODELS
from modelscope.utils.constant import Tasks

__all__ = ['WeNetAutomaticSpeechRecognition']


@MODELS.register_module(
    Tasks.auto_speech_recognition, module_name=Models.wenet_asr)
class WeNetAutomaticSpeechRecognition(Model):

    def __init__(self, model_dir: str, am_model_name: str,
                 model_config: Dict[str, Any], *args, **kwargs):
        """initialize the info of model.

        Args:
            model_dir (str): the model path.
        """
        super().__init__(model_dir, am_model_name, model_config, *args,
                         **kwargs)
        self.decoder = wenet.Decoder(model_dir, lang='chs')

    def forward(self, inputs: Dict[str, Any]) -> str:
        if inputs['audio_format'] == 'wav':
            rst = self.decoder.decode_wav(inputs['audio'])
        else:
            rst = self.decoder.decode(inputs['audio'])
        text = json.loads(rst)['nbest'][0]['sentence']
        return {'text': text}