语音流式识别 (ASR)

POST {{base_url}}/speech_to_text/v1/speech/stream_recognize

语音流式接口，将整个音频文件分片进行传入模型。能够实时返回数据。建议每个音频分片的大小为 100-200ms

参考接口文档：语音流式识别 (ASR)

Request Body

{"config"=>{"action"=>1, "engine_type"=>"16k_auto", "format"=>"pcm", "sequence_id"=>1, "stream_id"=>"asd1234567890ddd"}, "speech"=>{"speech"=>"PdmrfE267Cd/Z9KpmNFh71A2PSJZxSp7+8upCg=="}}