AWS SDKs を使用した文字起こし

import software.amazon.awssdk.auth.credentials.AwsCredentialsProvider;
import software.amazon.awssdk.auth.credentials.DefaultCredentialsProvider;
import software.amazon.awssdk.regions.Region;
import software.amazon.awssdk.services.transcribe.TranscribeClient;
import software.amazon.awssdk.services.transcribe.model.*;
import software.amazon.awssdk.services.transcribestreaming.model.LanguageCode;

public class TranscribeDemoApp {
    private static final Region REGION = Region.US_WEST_2;
    private static TranscribeClient client;

    public static void main(String args[]) {

        client = TranscribeClient.builder()
                .credentialsProvider(getCredentials())
                .region(REGION)
                .build();

        String transcriptionJobName = "my-first-transcription-job";
        String mediaType = "flac"; // can be other types
        Media myMedia = Media.builder()
                .mediaFileUri("s3://DOC-EXAMPLE-BUCKET/my-input-files/my-media-file.flac")
                .build();

        String outputS3BucketName = "s3://DOC-EXAMPLE-BUCKET";
        // Create the transcription job request
        StartTranscriptionJobRequest request = StartTranscriptionJobRequest.builder()
                .transcriptionJobName(transcriptionJobName)
                .languageCode(LanguageCode.EN_US.toString())
                .mediaSampleRateHertz(16000)
                .mediaFormat(mediaType)
                .media(myMedia)
                .outputBucketName(outputS3BucketName)
                .build();

        // send the request to start the transcription job
        StartTranscriptionJobResponse startJobResponse = client.startTranscriptionJob(request);

        System.out.println("Created the transcription job");
        System.out.println(startJobResponse.transcriptionJob());

        // Create the get job request
        GetTranscriptionJobRequest getJobRequest = GetTranscriptionJobRequest.builder()
                .transcriptionJobName(transcriptionJobName)
                .build();

        // send the request to get the transcription job including the job status
        GetTranscriptionJobResponse getJobResponse = client.getTranscriptionJob(getJobRequest);

        System.out.println("Get the transcription job request");
        System.out.println(getJobResponse.transcriptionJob());
    }

    private static AwsCredentialsProvider getCredentials() {
        return DefaultCredentialsProvider.create();
    }

}

const { TranscribeClient, StartTranscriptionJobCommand } = require("@aws-sdk/client-transcribe"); // CommonJS import

const region = "us-west-2";
const credentials = {
  "accessKeyId": "",
  "secretAccessKey": "",
};

const input = {
  TranscriptionJobName: "my-first-transcription-job",
  LanguageCode: "en-US",
  Media: {
    MediaFileUri: "s3://DOC-EXAMPLE-BUCKET/my-input-files/my-media-file.flac"
  },
  OutputBucketName: "DOC-EXAMPLE-BUCKET",
};

async function startTranscriptionRequest() {
  const transcribeConfig = {
    region,
    credentials
  };
  const transcribeClient = new TranscribeClient(transcribeConfig);
  const transcribeCommand = new StartTranscriptionJobCommand(input);
  try {
    const transcribeResponse = await transcribeClient.send(transcribeCommand);
    console.log("Transcription job created, the details:");
    console.log(transcribeResponse.TranscriptionJob);
  } catch(err) {
    console.log(err);
  }
}

startTranscriptionRequest();

import time
import boto3

def transcribe_file(job_name, file_uri, transcribe_client):
    transcribe_client.start_transcription_job(
        TranscriptionJobName = job_name,
        Media = {
            'MediaFileUri': file_uri
        },
        MediaFormat = 'flac',
        LanguageCode = 'en-US'
    )

    max_tries = 60
    while max_tries > 0:
        max_tries -= 1
        job = transcribe_client.get_transcription_job(TranscriptionJobName = job_name)
        job_status = job['TranscriptionJob']['TranscriptionJobStatus']
        if job_status in ['COMPLETED', 'FAILED']:
            print(f"Job {job_name} is {job_status}.")
            if job_status == 'COMPLETED':
                print(
                    f"Download the transcript from\n"
                    f"\t{job['TranscriptionJob']['Transcript']['TranscriptFileUri']}.")
            break
        else:
            print(f"Waiting for {job_name}. Current status is {job_status}.")
        time.sleep(10)


def main():
    transcribe_client = boto3.client('transcribe', region_name = 'us-west-2')
    file_uri = 's3://DOC-EXAMPLE-BUCKET/my-input-files/my-media-file.flac'
    transcribe_file('Example-job', file_uri, transcribe_client)


if __name__ == '__main__':
    main()

次の例は、音声ストリーミングを書き起こす Java プログラムです。

この例を実行するには、以下の点に注意します。

Amazon Transcribe 「ストリーミング用のクライアントを再試行する (Java SDK)」も参照してください。このコードは Amazon Transcribe への接続を管理し、接続にエラーがあるとデータの送信を再試行します。たとえば、ネットワーク上で一時的なエラーが発生した場合、このクライアントは失敗したリクエストを再送信します。

public class TranscribeStreamingDemoApp {
    private static final Region REGION = Region.US_WEST_2;
    private static TranscribeStreamingAsyncClient client;

    public static void main(String args[]) throws URISyntaxException, ExecutionException, InterruptedException, LineUnavailableException {

        client = TranscribeStreamingAsyncClient.builder()
                .credentialsProvider(getCredentials())
                .region(REGION)
                .build();

        CompletableFuture<Void> result = client.startStreamTranscription(getRequest(16_000),
                new AudioStreamPublisher(getStreamFromMic()),
                getResponseHandler());

        result.get();
        client.close();
    }

    private static InputStream getStreamFromMic() throws LineUnavailableException {

        // Signed PCM AudioFormat with 16,000 Hz, 16 bit sample size, mono
        int sampleRate = 16000;
        AudioFormat format = new AudioFormat(sampleRate, 16, 1, true, false);
        DataLine.Info info = new DataLine.Info(TargetDataLine.class, format);

        if (!AudioSystem.isLineSupported(info)) {
            System.out.println("Line not supported");
            System.exit(0);
        }

        TargetDataLine line = (TargetDataLine) AudioSystem.getLine(info);
        line.open(format);
        line.start();

        InputStream audioStream = new AudioInputStream(line);
        return audioStream;
    }

    private static AwsCredentialsProvider getCredentials() {
        return DefaultCredentialsProvider.create();
    }

    private static StartStreamTranscriptionRequest getRequest(Integer mediaSampleRateHertz) {
        return StartStreamTranscriptionRequest.builder()
                .languageCode(LanguageCode.EN_US.toString())
                .mediaEncoding(MediaEncoding.PCM)
                .mediaSampleRateHertz(mediaSampleRateHertz)
                .build();
    }

    private static StartStreamTranscriptionResponseHandler getResponseHandler() {
        return StartStreamTranscriptionResponseHandler.builder()
                .onResponse(r -> {
                    System.out.println("Received Initial response");
                })
                .onError(e -> {
                    System.out.println(e.getMessage());
                    StringWriter sw = new StringWriter();
                    e.printStackTrace(new PrintWriter(sw));
                    System.out.println("Error Occurred: " + sw.toString());
                })
                .onComplete(() -> {
                    System.out.println("=== All records stream successfully ===");
                })
                .subscriber(event -> {
                    List<Result> results = ((TranscriptEvent) event).transcript().results();
                    if (results.size() > 0) {
                        if (!results.get(0).alternatives().get(0).transcript().isEmpty()) {
                            System.out.println(results.get(0).alternatives().get(0).transcript());
                        }
                    }
                })
                .build();
    }

    private InputStream getStreamFromFile(String myMediaFileName) {
        try {
            File inputFile = new File(getClass().getClassLoader().getResource(myMediaFileName).getFile());
            InputStream audioStream = new FileInputStream(inputFile);
            return audioStream;
        } catch (FileNotFoundException e) {
            throw new RuntimeException(e);
        }
    }

    private static class AudioStreamPublisher implements Publisher<AudioStream> {
        private final InputStream inputStream;
        private static Subscription currentSubscription;


        private AudioStreamPublisher(InputStream inputStream) {
            this.inputStream = inputStream;
        }

        @Override
        public void subscribe(Subscriber<? super AudioStream> s) {

            if (this.currentSubscription == null) {
                this.currentSubscription = new SubscriptionImpl(s, inputStream);
            } else {
                this.currentSubscription.cancel();
                this.currentSubscription = new SubscriptionImpl(s, inputStream);
            }
            s.onSubscribe(currentSubscription);
        }
    }

    public static class SubscriptionImpl implements Subscription {
        private static final int CHUNK_SIZE_IN_BYTES = 1024 * 1;
        private final Subscriber<? super AudioStream> subscriber;
        private final InputStream inputStream;
        private ExecutorService executor = Executors.newFixedThreadPool(1);
        private AtomicLong demand = new AtomicLong(0);

        SubscriptionImpl(Subscriber<? super AudioStream> s, InputStream inputStream) {
            this.subscriber = s;
            this.inputStream = inputStream;
        }

        @Override
        public void request(long n) {
            if (n <= 0) {
                subscriber.onError(new IllegalArgumentException("Demand must be positive"));
            }

            demand.getAndAdd(n);

            executor.submit(() -> {
                try {
                    do {
                        ByteBuffer audioBuffer = getNextEvent();
                        if (audioBuffer.remaining() > 0) {
                            AudioEvent audioEvent = audioEventFromBuffer(audioBuffer);
                            subscriber.onNext(audioEvent);
                        } else {
                            subscriber.onComplete();
                            break;
                        }
                    } while (demand.decrementAndGet() > 0);
                } catch (Exception e) {
                    subscriber.onError(e);
                }
            });
        }

        @Override
        public void cancel() {
            executor.shutdown();
        }

        private ByteBuffer getNextEvent() {
            ByteBuffer audioBuffer = null;
            byte[] audioBytes = new byte[CHUNK_SIZE_IN_BYTES];

            int len = 0;
            try {
                len = inputStream.read(audioBytes);

                if (len <= 0) {
                    audioBuffer = ByteBuffer.allocate(0);
                } else {
                    audioBuffer = ByteBuffer.wrap(audioBytes, 0, len);
                }
            } catch (IOException e) {
                throw new UncheckedIOException(e);
            }

            return audioBuffer;
        }

        private AudioEvent audioEventFromBuffer(ByteBuffer bb) {
            return AudioEvent.builder()
                    .audioChunk(SdkBytes.fromByteBuffer(bb))
                    .build();
        }
    }
}

const {
  TranscribeStreamingClient,
  StartStreamTranscriptionCommand,
} = require("@aws-sdk/client-transcribe-streaming");
const { createReadStream } = require("fs");
const { join } = require("path");

const audio = createReadStream(join(__dirname, "my-media-file.flac"), { highWaterMark: 1024 * 16});

const LanguageCode = "en-US";
const MediaEncoding = "pcm";
const MediaSampleRateHertz = "16000";
const credentials = {
  "accessKeyId": "",
  "secretAccessKey": "",
};
async function startRequest() {
  const client = new TranscribeStreamingClient({
    region: "us-west-2",
    credentials
  });

  const params = {
    LanguageCode,
    MediaEncoding,
    MediaSampleRateHertz,
    AudioStream: (async function* () {
      for await (const chunk of audio) {
        yield {AudioEvent: {AudioChunk: chunk}};
      }
    })(),
  };
  const command = new StartStreamTranscriptionCommand(params);
  // Send transcription request
  const response = await client.send(command);
  // Start to print response
  try {
    for await (const event of response.TranscriptResultStream) {
      console.log(JSON.stringify(event));
    }
  } catch(err) {
    console.log("error")
    console.log(err)
  }
}
startRequest();

次の例は、音声ストリーミングを書き起こす Python プログラムです。

この例を実行するには、以下の点に注意します。

import asyncio
# This example uses aiofile for asynchronous file reads.
# It's not a dependency of the project but can be installed
# with `pip install aiofile`.
import aiofile

from amazon_transcribe.client import TranscribeStreamingClient
from amazon_transcribe.handlers import TranscriptResultStreamHandler
from amazon_transcribe.model import TranscriptEvent

"""
Here's an example of a custom event handler you can extend to
process the returned transcription results as needed. This
handler will simply print the text out to your interpreter.
"""
class MyEventHandler(TranscriptResultStreamHandler):
    async def handle_transcript_event(self, transcript_event: TranscriptEvent):
        # This handler can be implemented to handle transcriptions as needed.
        # Here's an example to get started.
        results = transcript_event.transcript.results
        for result in results:
            for alt in result.alternatives:
                print(alt.transcript)


async def basic_transcribe():
    # Set up our client with your chosen Region
    client = TranscribeStreamingClient(region = "us-west-2")

    # Start transcription to generate async stream
    stream = await client.start_stream_transcription(
        language_code = "en-US",
        media_sample_rate_hz = 16000,
        media_encoding = "pcm",
    )

    async def write_chunks():
        # NOTE: For pre-recorded files longer than 5 minutes, the sent audio
        # chunks should be rate limited to match the real-time bitrate of the
        # audio stream to avoid signing issues.
        async with aiofile.AIOFile('filepath/my-media-file.flac', 'rb') as afp:
            reader = aiofile.Reader(afp, chunk_size = 1024 * 16)
            async for chunk in reader:
                await stream.input_stream.send_audio_event(audio_chunk = chunk)
        await stream.input_stream.end_stream()

    # Instantiate our handler and start processing events
    handler = MyEventHandler(stream.output_stream)
    await asyncio.gather(write_chunks(), handler.handle_events())

loop = asyncio.get_event_loop()
loop.run_until_complete(basic_transcribe())
loop.close()

ストリーミング C++ SDK の例については、コード例の章を参照してください。