1

我们在使用一个号码的 Google 语音转文本服务方面遇到了一些问题(直到现在):

数字以 ...35984778 结尾,我们正在变成 ...35984 526。最后 3 个数字完全是错误的。口语是德语。

可以通过以下步骤重现该问题:

  • 用户通过电话提供他的电话号码 ...35984778。此数据被发送到 Google 语音转文本 API。
  • Google 语音转文本服务提供 ...35984 526

其他数字与 ...09778 之类的结尾一起正常工作。

以下是应用程序日志文件中的一些行:

...
{"pid":12,"hostname":"","level":20,"time":1588397981946,"msg":"starting with the following Google stream recognition options:","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"Goo gleRecognizeStream","config":{"encoding":"MULAW","sampleRateHertz":8000,"languageCode":"de-DE","profanityFilter":true,"speechContexts":[{"phrases":["1","2","3","4","5","6","7","8","9","0"]}]},"singleUtterance":true,"interimResults ":true,"verbose":true,"v":1} {"pid":12,"hostname":"","level":20,"time":1588397981946,"msg":"starting session with sessionID 8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"SpeechToTextAdapter","v":1} {"pid":12,"hostname":"","level":30,"time":1588397981947,"msg":"...: The speech recognition session started.","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","v":1} {"pid":12,"hostname":"","level":20,"time":1588397985083,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397985278,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397985677,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397985877,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397985982,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397986383,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397986986,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397987085,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397987481,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397987778,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397988381,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397988577,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397988674,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397988680,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397989275,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397989875,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397990375,"msg":"received interim result","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":30,"time":1588397990947,"msg":"...: Received a transcription with a code unit length of 13 and a confidence score of 0.9190642237663269","name":"SpeechToTextAdapter","sessionID":" 8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397990948,"msg":"onRecognitionStop","name":"SpeechToTextAdapter","v":1} {"pid":12,"hostname":"","level":30,"time":1588397990948,"msg":"...: The speech recognition session ended.","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","v":1} {"pid":12,"hostname":"","level":20,"time":1588397990948,"msg":"ended","name":"SpeechToTextAdapter","sessionID":"8a3f07ca-4351-4e5d-b8c8-bcddc10468f9","name":"GoogleRecognizeStream","v":1} {"pid":12,"hostname":"","level":20,"time":1588397990952,"msg":"session closed","name":"SpeechToTextAdapter","v":1} {"pid":12,"hostname":"","level":30,"time":1588397990954,"msg":"...: Received a request to start a speech recognition session.","name":"SpeechToTextAdapter","v":1} {"pid":12,"hostname":"","level":20,"time":1588397995351,"msg":"session closed","name":"SpeechToTextAdapter","v":1}
...

这个问题也可以用安卓手机重现。我的一位同事用她的手机进行了测试,并且确实出现了同样的问题。此测试仅在手机上进行,而不是在集成了 Google 语音转文本服务的应用程序上进行(如上所述)。

4

0 回答 0