[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-08-18。"],[],[],null,["# Select a transcription model\n\nThis page describes how to use a specific machine learning model for\naudio transcription requests to Speech-to-Text.\n\nTranscription models\n--------------------\n\nSpeech-to-Text detects words in an audio clip by comparing input\nto one of many machine learning *models*. Each model has been trained\nby analyzing millions of examples---in this case, many, many audio\nrecordings of people speaking.\n\nSpeech-to-Text has specialized models which are trained from audio\nfor specific sources. These models provide better results\nwhen applied toward similar kinds of audio data to the data they were trained\non.\n\nFor example, Speech-to-Text has a transcription model trained\nto recognize speech recorded over the phone. When Speech-to-Text\nuses the `telephony_short` or `telephony` model to transcribe phone audio, it\nproduces more accurate transcription results than if it had transcribed phone\naudio using the `short` or `long` models.\n\nThe following table shows the transcription models available for\nuse with Speech-to-Text.\n\nSelect a model for audio transcription\n--------------------------------------\n\nThe model is specified by the [Recognizer](/speech-to-text/v2/docs/reference/rest/v2/projects.locations.recognizers#Recognizer)\nused for the recognition request. Call [`speech/projects.locations.recognizers/create`](/speech-to-text/v2/docs/reference/rest/v2/projects.locations.recognizers/create)\nto create a recognizer, and use the `model` field to specify the model. Valid\nmodels for each language can be found in the [Supported Languages](/speech-to-text/v2/docs/speech-to-text-supported-languages) table."]]