Google Speech Translation in Salesforce Apex

AudioData.apxc

public class AudioData {
	public class Audio {
		public String content;
	}
	public class Config {
		public String encoding;
		public Integer sampleRateHertz;
		public String languageCode;
		public Boolean enableWordTimeOffsets;
	}
	public Config config;
	public Audio audio;
	
	public static AudioData parse(String json) {
		return (AudioData) System.JSON.deserialize(json, AudioData.class);
	}
}

AudioResponseData.apxc

public class AudioResponseData {
    public class SpeechRecognitionResult
    {
        public List<SpeechRecognitionAlternative> alternatives;
        public string channelTag;
  public string languageCode;
    }
    public class SpeechRecognitionAlternative
    {
        public string transcript;
        public string confidence;
        public List<WordInfo> words;
    }
    public class WordInfo
    {
        public string startTime;
        public string endTime;
        public string word;
        public string confidence;
        public string speakerTag;
    }
   
    public List<SpeechRecognitionResult> results;
   
    public static AudioResponseData parse(String json) {
return (AudioResponseData) System.JSON.deserialize(json, AudioResponseData.class);
}
}

GoogleSpeechService.apxc

public class GoogleSpeechService {
    
   	public static string getAccessToken()
    {
        GoogleAuthSetting__mdt mapping = 
		[SELECT AccessToken__c, Label FROM GoogleAuthSetting__mdt WHERE Label='AccessToken' and DeveloperName='AccessToken'];
        return mapping.AccessToken__c;
    }
    
    @AuraEnabled
    public static string GetTranscript(string recordid)
    {
        List<ContentDocumentLink> links=[SELECT ContentDocumentId,LinkedEntityId FROM ContentDocumentLink where LinkedEntityId=:recordId];
        Set<Id> ids=new Set<Id>();
        for(ContentDocumentLink link:links)
        {
            ids.add(link.ContentDocumentId);
        }
        List<ContentVersion> versions=[SELECT VersionData,Title,ContentDocumentId,FileExtension FROM ContentVersion WHERE ContentDocumentId = :ids AND IsLatest = true];
        blob file_body;
        if(versions!=null && versions.size()>0)
        {
            file_body=versions[0].VersionData;
        }
        
        HttpRequest req = new HttpRequest();
        req.setHeader('Authorization','Bearer '+ getAccessToken());
        req.setHeader('Content-Type','application/json; charset=utf-8');
        req.setMethod('POST');
        req.setEndpoint('https://speech.googleapis.com/v1p1beta1/speech:recognize');
        req.setTimeout(120000);
        
        AudioData data=new AudioData();
        AudioData.Audio audio=new AudioData.Audio();
        audio.content=EncodingUtil.base64Encode(file_body);
        data.audio=audio;
        AudioData.Config config=new AudioData.Config();
        config.encoding='MP3';
        config.sampleRateHertz=16000;
        config.languageCode='en-US';
        config.enableWordTimeOffsets=false;
        data.config=config;
        string jsondta=system.JSON.serialize(data);
        system.debug('jsondta:'+jsondta);
        req.setBody(jsondta);
        Http http = new Http();
        HTTPResponse res = http.send(req);
        AudioResponseData responseData=AudioResponseData.parse(res.getBody());
        system.debug('res:'+res.getBody());
      
        string content='';
        for(AudioResponseData.SpeechRecognitionResult recognitionResult:responseData.results)
        {
            for(AudioResponseData.SpeechRecognitionAlternative alternative: recognitionResult.alternatives)
            {
               content=content+'\n'+alternative.transcript;
            }
        }
        
       	CaseComment newCommmand = new CaseComment();
        newCommmand.CommentBody = content;
        newCommmand.IsPublished = TRUE;
        newCommmand.ParentId = recordid;
        insert newCommmand;
        system.debug('res:'+newCommmand.id);
        return newCommmand.id;
    } 
}