긴 오디오 파일을 텍스트로 변환

이 페이지에서는 비동기 음성 인식을 사용하여 긴 오디오 파일(1분 이상)을 텍스트로 변환하는 방법을 설명합니다.

비동기 음성 인식은 장기 실행 오디오 처리 작업을 시작합니다. 비동기 음성 인식을 사용하여 1분 이상의 오디오를 인식할 수 있습니다. 이보다 짧은 오디오는 동기 음성 인식으로 더 빠르고 단순하게 처리할 수 있습니다.

google.longrunning.Operations 인터페이스를 통해 작업 결과를 검색할 수 있습니다. 결과는 5일(120시간) 동안 검색할 수 있습니다. 오디오 콘텐츠를 Speech-to-Text에 직접 보내거나 Speech-to-Text가 Google Cloud Storage에 있는 기존 오디오 콘텐츠를 처리할 수 있습니다. 비동기 음성 인식 요청에 대한 오디오 제한도 참조하세요.

Speech-to-Text v1은 공식 출시되었으며 일반적으로 https://speech.googleapis.com/v1/speech 엔드포인트에서 사용할 수 있습니다. 클라이언트 라이브러리는 알파 버전으로 출시되었으며 이전 버전과 호환되지 않는 방식으로 변경될 수 있습니다. 현재는 클라이언트 라이브러리를 프로덕션 용도로 사용하지 않는 것이 좋습니다.

이 샘플을 사용하려면 gcloud를 설정하고 서비스 계정을 만들어 활성화해야 합니다. gcloud를 설정하고 서비스 계정을 만들어 활성화하는 방법은 빠른 시작을 참조하세요.

Google Cloud Storage 파일을 사용하여 긴 오디오 파일 텍스트 변환

이 샘플은 Cloud Storage 버킷을 사용하여 장기 실행 텍스트 변환 프로세스의 원시 오디오 입력을 저장합니다.

프로토콜

자세한 내용은 speech:longrunningrecognize API 엔드포인트를 참조하세요.

동기 음성 인식을 수행하려면 POST 요청을 실행하고 적절한 요청 본문을 제공합니다. 다음은 curl을 사용한 POST 요청의 예시입니다. 이 예시에서는 Google Cloud Cloud SDK를 사용하는 프로젝트용으로 설정된 서비스 계정의 액세스 토큰을 사용합니다. Cloud SDK를 설치하고, 서비스 계정으로 프로젝트를 설정하고, 액세스 토큰을 획득하는 방법은 빠른 시작을 참조하세요.

    curl -X POST \
         -H "Authorization: Bearer "$(gcloud auth application-default print-access-token) \
         -H "Content-Type: application/json; charset=utf-8" \
         --data "{
      'config': {
        'language_code': 'en-US'
      },
      'audio':{
        'uri':'gs://gcs-test-data/vr.flac'
      }
    }" "https://speech.googleapis.com/v1/speech:longrunningrecognize"
    

요청 본문 구성에 대한 자세한 내용은 RecognitionConfigRecognitionAudio 참조 문서를 확인하세요.

요청이 성공하면 서버가 200 OK HTTP 상태 코드와 응답을 JSON 형식으로 반환합니다.

    {
      "name": "7612202767953098924"
    }
    

여기서 name은 요청에 대해 생성된 장기 실행 작업의 이름입니다.

처리가 완료될 때까지 기다립니다. 처리 시간은 소스 오디오에 따라 다르며, 대부분의 경우 소스 오디오 길이의 절반 정도 시간이면 결과를 얻을 수 있습니다. 장기 실행 작업의 상태는 https://speech.googleapis.com/v1/operations/ 엔드포인트에 대해 GET 요청을 실행하여 알아볼 수 있습니다. 이때 your-operation-namelongrunningrecognize 요청에서 반환된 name으로 바꿉니다. 요청의 예상 진행률은 progressPercent 필드에서 확인할 수 있습니다.

    curl -H "Authorization: Bearer "$(gcloud auth application-default print-access-token) \
         -H "Content-Type: application/json; charset=utf-8" \
         "https://speech.googleapis.com/v1/operations/your-operation-name"
    

요청이 성공하면 서버가 200 OK HTTP 상태 코드와 응답을 JSON 형식으로 반환합니다.

    {
      "name": "7612202767953098924",
      "metadata": {
        "@type": "type.googleapis.com/google.cloud.speech.v1.LongRunningRecognizeMetadata",
        "progressPercent": 100,
        "startTime": "2017-07-20T16:36:55.033650Z",
        "lastUpdateTime": "2017-07-20T16:37:17.158630Z"
      },
      "done": true,
      "response": {
        "@type": "type.googleapis.com/google.cloud.speech.v1.LongRunningRecognizeResponse",
        "results": [
          {
            "alternatives": [
              {
                "transcript": "okay so what am I doing here...(etc)...",
                "confidence": 0.96096134,
              }
            ]
          },
          {
            "alternatives": [
              {
                ...
              }
            ]
          }
        ]
      }
    }
    

작업이 완료되지 않았으면 응답의 done 속성이 true가 될 때까지 GET 요청을 반복해서 엔드포인트를 폴링할 수 있습니다.

gcloud 명령어

자세한 내용은 recognize-long-running 명령어를 참조하세요.

비동기 음성 인식을 수행하려면 gcloud 명령줄 도구를 사용하여 로컬 파일의 경로나 Google Cloud Storage URL을 제공합니다.

    gcloud ml speech recognize-long-running \
        'gs://cloud-samples-tests/speech/brooklyn.flac' \
         --language-code='en-US' --async
    

요청이 성공하면 서버가 장기 실행 작업의 ID를 JSON 형식으로 반환합니다.

    {
      "name": OPERATION_ID
    }
    

그러면 다음 명령어를 실행하여 작업에 대한 정보를 얻을 수 있습니다.

    gcloud ml speech operations describe OPERATION_ID
    

또한 다음 명령어를 실행하여 작업이 완료될 때까지 작업을 폴링할 수 있습니다.

    gcloud ml speech operations wait OPERATION_ID
    

작업이 완료되면 오디오가 JSON 형식으로 반환됩니다.

    {
      "@type": "type.googleapis.com/google.cloud.speech.v1.LongRunningRecognizeResponse",
      "results": [
        {
          "alternatives": [
            {
              "confidence": 0.9840146,
              "transcript": "how old is the Brooklyn Bridge"
            }
          ]
        }
      ]
    }
    

C#

static object AsyncRecognizeGcs(string storageUri)
    {
        var speech = SpeechClient.Create();
        var longOperation = speech.LongRunningRecognize(new RecognitionConfig()
        {
            Encoding = RecognitionConfig.Types.AudioEncoding.Linear16,
            SampleRateHertz = 16000,
            LanguageCode = "en",
        }, RecognitionAudio.FromStorageUri(storageUri));
        longOperation = longOperation.PollUntilCompleted();
        var response = longOperation.Result;
        foreach (var result in response.Results)
        {
            foreach (var alternative in result.Alternatives)
            {
                Console.WriteLine($"Transcript: { alternative.Transcript}");
            }
        }
        return 0;
    }

Go


    func sendGCS(w io.Writer, client *speech.Client, gcsURI string) error {
    	ctx := context.Background()

    	// Send the contents of the audio file with the encoding and
    	// and sample rate information to be transcripted.
    	req := &speechpb.LongRunningRecognizeRequest{
    		Config: &speechpb.RecognitionConfig{
    			Encoding:        speechpb.RecognitionConfig_LINEAR16,
    			SampleRateHertz: 16000,
    			LanguageCode:    "en-US",
    		},
    		Audio: &speechpb.RecognitionAudio{
    			AudioSource: &speechpb.RecognitionAudio_Uri{Uri: gcsURI},
    		},
    	}

    	op, err := client.LongRunningRecognize(ctx, req)
    	if err != nil {
    		return err
    	}
    	resp, err := op.Wait(ctx)
    	if err != nil {
    		return err
    	}

    	// Print the results.
    	for _, result := range resp.Results {
    		for _, alt := range result.Alternatives {
    			fmt.Fprintf(w, "\"%v\" (confidence=%3f)\n", alt.Transcript, alt.Confidence)
    		}
    	}
    	return nil
    }
    

자바

/**
     * Performs non-blocking speech recognition on remote FLAC file and prints the transcription.
     *
     * @param gcsUri the path to the remote LINEAR16 audio file to transcribe.
     */
    public static void asyncRecognizeGcs(String gcsUri) throws Exception {
      // Instantiates a client with GOOGLE_APPLICATION_CREDENTIALS
      try (SpeechClient speech = SpeechClient.create()) {

        // Configure remote file request for FLAC
        RecognitionConfig config =
            RecognitionConfig.newBuilder()
                .setEncoding(AudioEncoding.FLAC)
                .setLanguageCode("en-US")
                .setSampleRateHertz(16000)
                .build();
        RecognitionAudio audio = RecognitionAudio.newBuilder().setUri(gcsUri).build();

        // Use non-blocking call for getting file transcription
        OperationFuture<LongRunningRecognizeResponse, LongRunningRecognizeMetadata> response =
            speech.longRunningRecognizeAsync(config, audio);
        while (!response.isDone()) {
          System.out.println("Waiting for response...");
          Thread.sleep(10000);
        }

        List<SpeechRecognitionResult> results = response.get().getResultsList();

        for (SpeechRecognitionResult result : results) {
          // There can be several alternative transcripts for a given chunk of speech. Just use the
          // first (most likely) one here.
          SpeechRecognitionAlternative alternative = result.getAlternativesList().get(0);
          System.out.printf("Transcription: %s\n", alternative.getTranscript());
        }
      }
    }

Node.js

// Imports the Google Cloud client library
    const speech = require('@google-cloud/speech');

    // Creates a client
    const client = new speech.SpeechClient();

    /**
     * TODO(developer): Uncomment the following lines before running the sample.
     */
    // const gcsUri = 'gs://my-bucket/audio.raw';
    // const encoding = 'Encoding of the audio file, e.g. LINEAR16';
    // const sampleRateHertz = 16000;
    // const languageCode = 'BCP-47 language code, e.g. en-US';

    const config = {
      encoding: encoding,
      sampleRateHertz: sampleRateHertz,
      languageCode: languageCode,
    };

    const audio = {
      uri: gcsUri,
    };

    const request = {
      config: config,
      audio: audio,
    };

    // Detects speech in the audio file. This creates a recognition job that you
    // can wait for now, or get its result later.
    const [operation] = await client.longRunningRecognize(request);
    // Get a Promise representation of the final result of the job
    const [response] = await operation.promise();
    const transcription = response.results
      .map(result => result.alternatives[0].transcript)
      .join('\n');
    console.log(`Transcription: ${transcription}`);

PHP

use Google\Cloud\Speech\V1\SpeechClient;
    use Google\Cloud\Speech\V1\RecognitionAudio;
    use Google\Cloud\Speech\V1\RecognitionConfig;
    use Google\Cloud\Speech\V1\RecognitionConfig\AudioEncoding;

    /** Uncomment and populate these variables in your code */
    // $uri = 'The Cloud Storage object to transcribe (gs://your-bucket-name/your-object-name)';

    // change these variables if necessary
    $encoding = AudioEncoding::LINEAR16;
    $sampleRateHertz = 32000;
    $languageCode = 'en-US';

    // set string as audio content
    $audio = (new RecognitionAudio())
        ->setUri($uri);

    // set config
    $config = (new RecognitionConfig())
        ->setEncoding($encoding)
        ->setSampleRateHertz($sampleRateHertz)
        ->setLanguageCode($languageCode);

    // create the speech client
    $client = new SpeechClient();

    // create the asyncronous recognize operation
    $operation = $client->longRunningRecognize($config, $audio);
    $operation->pollUntilComplete();

    if ($operation->operationSucceeded()) {
        $response = $operation->getResult();

        // each result is for a consecutive portion of the audio. iterate
        // through them to get the transcripts for the entire audio file.
        foreach ($response->getResults() as $result) {
            $alternatives = $result->getAlternatives();
            $mostLikely = $alternatives[0];
            $transcript = $mostLikely->getTranscript();
            $confidence = $mostLikely->getConfidence();
            printf('Transcript: %s' . PHP_EOL, $transcript);
            printf('Confidence: %s' . PHP_EOL, $confidence);
        }
    } else {
        print_r($operation->getError());
    }

    $client->close();

Python

from google.cloud import speech_v1
    from google.cloud.speech_v1 import enums

    def sample_long_running_recognize(storage_uri):
        """
        Transcribe long audio file from Cloud Storage using asynchronous speech
        recognition

        Args:
          storage_uri URI for audio file in Cloud Storage, e.g. gs://[BUCKET]/[FILE]
        """

        client = speech_v1.SpeechClient()

        # storage_uri = 'gs://cloud-samples-data/speech/brooklyn_bridge.raw'

        # Sample rate in Hertz of the audio data sent
        sample_rate_hertz = 16000

        # The language of the supplied audio
        language_code = "en-US"

        # Encoding of audio data sent. This sample sets this explicitly.
        # This field is optional for FLAC and WAV audio formats.
        encoding = enums.RecognitionConfig.AudioEncoding.LINEAR16
        config = {
            "sample_rate_hertz": sample_rate_hertz,
            "language_code": language_code,
            "encoding": encoding,
        }
        audio = {"uri": storage_uri}

        operation = client.long_running_recognize(config, audio)

        print(u"Waiting for operation to complete...")
        response = operation.result()

        for result in response.results:
            # First alternative is the most probable result
            alternative = result.alternatives[0]
            print(u"Transcript: {}".format(alternative.transcript))

    

Ruby

# storage_path = "Path to file in Cloud Storage, eg. gs://bucket/audio.raw"

    require "google/cloud/speech"

    speech = Google::Cloud::Speech.new

    config = { encoding:          :LINEAR16,
               sample_rate_hertz: 16_000,
               language_code:     "en-US" }
    audio = { uri: storage_path }

    operation = speech.long_running_recognize config, audio

    puts "Operation started"

    operation.wait_until_done!

    raise operation.results.message if operation.error?

    results = operation.response.results

    alternatives = results.first.alternatives
    alternatives.each do |alternative|
      puts "Transcription: #{alternative.transcript}"
    end

로컬 파일을 사용하여 긴 오디오 파일 텍스트 변환

이 샘플은 로컬 파일을 사용하여 장기 실행 텍스트 변환 프로세스의 원시 오디오 입력을 저장합니다.

C#

static object LongRunningRecognize(string filePath)
    {
        var speech = SpeechClient.Create();
        var longOperation = speech.LongRunningRecognize(new RecognitionConfig()
        {
            Encoding = RecognitionConfig.Types.AudioEncoding.Linear16,
            SampleRateHertz = 16000,
            LanguageCode = "en",
        }, RecognitionAudio.FromFile(filePath));
        longOperation = longOperation.PollUntilCompleted();
        var response = longOperation.Result;
        foreach (var result in response.Results)
        {
            foreach (var alternative in result.Alternatives)
            {
                Console.WriteLine(alternative.Transcript);
            }
        }
        return 0;
    }

Go


    func send(w io.Writer, client *speech.Client, filename string) error {
    	ctx := context.Background()
    	data, err := ioutil.ReadFile(filename)
    	if err != nil {
    		return err
    	}

    	// Send the contents of the audio file with the encoding and
    	// and sample rate information to be transcripted.
    	req := &speechpb.LongRunningRecognizeRequest{
    		Config: &speechpb.RecognitionConfig{
    			Encoding:        speechpb.RecognitionConfig_LINEAR16,
    			SampleRateHertz: 16000,
    			LanguageCode:    "en-US",
    		},
    		Audio: &speechpb.RecognitionAudio{
    			AudioSource: &speechpb.RecognitionAudio_Content{Content: data},
    		},
    	}

    	op, err := client.LongRunningRecognize(ctx, req)
    	if err != nil {
    		return err
    	}
    	resp, err := op.Wait(ctx)
    	if err != nil {
    		return err
    	}

    	// Print the results.
    	for _, result := range resp.Results {
    		for _, alt := range result.Alternatives {
    			fmt.Fprintf(w, "\"%v\" (confidence=%3f)\n", alt.Transcript, alt.Confidence)
    		}
    	}
    	return nil
    }
    

자바

/**
     * Performs non-blocking speech recognition on raw PCM audio and prints the transcription. Note
     * that transcription is limited to 60 seconds audio.
     *
     * @param fileName the path to a PCM audio file to transcribe.
     */
    public static void asyncRecognizeFile(String fileName) throws Exception {
      // Instantiates a client with GOOGLE_APPLICATION_CREDENTIALS
      try (SpeechClient speech = SpeechClient.create()) {

        Path path = Paths.get(fileName);
        byte[] data = Files.readAllBytes(path);
        ByteString audioBytes = ByteString.copyFrom(data);

        // Configure request with local raw PCM audio
        RecognitionConfig config =
            RecognitionConfig.newBuilder()
                .setEncoding(AudioEncoding.LINEAR16)
                .setLanguageCode("en-US")
                .setSampleRateHertz(16000)
                .build();
        RecognitionAudio audio = RecognitionAudio.newBuilder().setContent(audioBytes).build();

        // Use non-blocking call for getting file transcription
        OperationFuture<LongRunningRecognizeResponse, LongRunningRecognizeMetadata> response =
            speech.longRunningRecognizeAsync(config, audio);

        while (!response.isDone()) {
          System.out.println("Waiting for response...");
          Thread.sleep(10000);
        }

        List<SpeechRecognitionResult> results = response.get().getResultsList();

        for (SpeechRecognitionResult result : results) {
          // There can be several alternative transcripts for a given chunk of speech. Just use the
          // first (most likely) one here.
          SpeechRecognitionAlternative alternative = result.getAlternativesList().get(0);
          System.out.printf("Transcription: %s%n", alternative.getTranscript());
        }
      }
    }

Node.js

// Imports the Google Cloud client library
    const speech = require('@google-cloud/speech');
    const fs = require('fs');

    // Creates a client
    const client = new speech.SpeechClient();

    /**
     * TODO(developer): Uncomment the following lines before running the sample.
     */
    // const filename = 'Local path to audio file, e.g. /path/to/audio.raw';
    // const encoding = 'Encoding of the audio file, e.g. LINEAR16';
    // const sampleRateHertz = 16000;
    // const languageCode = 'BCP-47 language code, e.g. en-US';

    const config = {
      encoding: encoding,
      sampleRateHertz: sampleRateHertz,
      languageCode: languageCode,
    };
    const audio = {
      content: fs.readFileSync(filename).toString('base64'),
    };

    const request = {
      config: config,
      audio: audio,
    };

    // Detects speech in the audio file. This creates a recognition job that you
    // can wait for now, or get its result later.
    const [operation] = await client.longRunningRecognize(request);

    // Get a Promise representation of the final result of the job
    const [response] = await operation.promise();
    const transcription = response.results
      .map(result => result.alternatives[0].transcript)
      .join('\n');
    console.log(`Transcription: ${transcription}`);

PHP

use Google\Cloud\Speech\V1\SpeechClient;
    use Google\Cloud\Speech\V1\RecognitionAudio;
    use Google\Cloud\Speech\V1\RecognitionConfig;
    use Google\Cloud\Speech\V1\RecognitionConfig\AudioEncoding;

    /** Uncomment and populate these variables in your code */
    // $audioFile = 'path to an audio file';

    // change these variables if necessary
    $encoding = AudioEncoding::LINEAR16;
    $sampleRateHertz = 32000;
    $languageCode = 'en-US';

    // get contents of a file into a string
    $content = file_get_contents($audioFile);

    // set string as audio content
    $audio = (new RecognitionAudio())
        ->setContent($content);

    // set config
    $config = (new RecognitionConfig())
        ->setEncoding($encoding)
        ->setSampleRateHertz($sampleRateHertz)
        ->setLanguageCode($languageCode);

    // create the speech client
    $client = new SpeechClient();

    // create the asyncronous recognize operation
    $operation = $client->longRunningRecognize($config, $audio);
    $operation->pollUntilComplete();

    if ($operation->operationSucceeded()) {
        $response = $operation->getResult();

        // each result is for a consecutive portion of the audio. iterate
        // through them to get the transcripts for the entire audio file.
        foreach ($response->getResults() as $result) {
            $alternatives = $result->getAlternatives();
            $mostLikely = $alternatives[0];
            $transcript = $mostLikely->getTranscript();
            $confidence = $mostLikely->getConfidence();
            printf('Transcript: %s' . PHP_EOL, $transcript);
            printf('Confidence: %s' . PHP_EOL, $confidence);
        }
    } else {
        print_r($operation->getError());
    }

    $client->close();

Python

from google.cloud import speech_v1
    from google.cloud.speech_v1 import enums
    import io

    def sample_long_running_recognize(local_file_path):
        """
        Transcribe a long audio file using asynchronous speech recognition

        Args:
          local_file_path Path to local audio file, e.g. /path/audio.wav
        """

        client = speech_v1.SpeechClient()

        # local_file_path = 'resources/brooklyn_bridge.raw'

        # The language of the supplied audio
        language_code = "en-US"

        # Sample rate in Hertz of the audio data sent
        sample_rate_hertz = 16000

        # Encoding of audio data sent. This sample sets this explicitly.
        # This field is optional for FLAC and WAV audio formats.
        encoding = enums.RecognitionConfig.AudioEncoding.LINEAR16
        config = {
            "language_code": language_code,
            "sample_rate_hertz": sample_rate_hertz,
            "encoding": encoding,
        }
        with io.open(local_file_path, "rb") as f:
            content = f.read()
        audio = {"content": content}

        operation = client.long_running_recognize(config, audio)

        print(u"Waiting for operation to complete...")
        response = operation.result()

        for result in response.results:
            # First alternative is the most probable result
            alternative = result.alternatives[0]
            print(u"Transcript: {}".format(alternative.transcript))

    

Ruby

# audio_file_path = "Path to file on which to perform speech recognition"

    require "google/cloud/speech"

    speech = Google::Cloud::Speech.new

    audio_file = File.binread audio_file_path
    config     = { encoding:          :LINEAR16,
                   sample_rate_hertz: 16_000,
                   language_code:     "en-US" }
    audio      = { content: audio_file }

    operation = speech.long_running_recognize config, audio

    puts "Operation started"

    operation.wait_until_done!

    raise operation.results.message if operation.error?

    results = operation.response.results

    alternatives = results.first.alternatives
    alternatives.each do |alternative|
      puts "Transcription: #{alternative.transcript}"
    end