Package com.composum.ai.backend.base.service.chat.impl

Class GPTDictationServiceImpl

java.lang.Object

com.composum.ai.backend.base.service.chat.impl.GPTDictationServiceImpl

All Implemented Interfaces:

GPTDictationService

public class GPTDictationServiceImpl
extends Object
implements GPTDictationService

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static interface	GPTDictationServiceImpl.GPTDictationServiceConfig	Configures whether it's enabled (default false), the model and the request counts, and the maximum request size.
protected class	GPTDictationServiceImpl.LimitedInputStream

Field Summary

Fields

Modifier and Type	Field	Description
protected static int	DEFAULT_MAX_REQUEST_SIZE
protected static String	DEFAULT_MODEL
protected static int	DEFAULTVALUE_REQUESTS_PER_DAY
protected static int	DEFAULTVALUE_REQUESTS_PER_HOUR
protected static int	DEFAULTVALUE_REQUESTS_PER_MINUTE
protected boolean	enabled
protected org.apache.hc.client5.http.impl.classic.CloseableHttpClient	httpClient
protected RateLimiter	limiter
protected long	maxRequestSize
protected String	model
protected GPTInternalOpenAIHelper	openAIHelper
static String	URL_OPENAI_TRANSCRIPTIONS

Constructor Summary

Constructors

Constructor	Description
GPTDictationServiceImpl()

Method Summary

Modifier and Type	Method	Description
protected void	activate(GPTDictationServiceImpl.GPTDictationServiceConfig config)
protected void	deactivate()
boolean	isAvailable(GPTConfiguration configuration)	Whether the service is enabled and properly configured.
protected void	modified(GPTDictationServiceImpl.GPTDictationServiceConfig config)
String	transcribe(InputStream audioStream, String contentType, String language, GPTConfiguration configuration, String prompt)	Transcribes the input audio to text.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

URL_OPENAI_TRANSCRIPTIONS

public static final String URL_OPENAI_TRANSCRIPTIONS

See Also:

Constant Field Values

DEFAULTVALUE_REQUESTS_PER_MINUTE

protected static final int DEFAULTVALUE_REQUESTS_PER_MINUTE

See Also:

Constant Field Values

DEFAULTVALUE_REQUESTS_PER_HOUR

protected static final int DEFAULTVALUE_REQUESTS_PER_HOUR

See Also:

Constant Field Values

DEFAULTVALUE_REQUESTS_PER_DAY

protected static final int DEFAULTVALUE_REQUESTS_PER_DAY

See Also:

Constant Field Values

DEFAULT_MODEL

protected static final String DEFAULT_MODEL

See Also:

Constant Field Values

DEFAULT_MAX_REQUEST_SIZE

protected static final int DEFAULT_MAX_REQUEST_SIZE

See Also:

Constant Field Values

httpClient

protected org.apache.hc.client5.http.impl.classic.CloseableHttpClient httpClient

limiter

protected RateLimiter limiter

enabled

protected boolean enabled

model

protected String model

maxRequestSize

protected long maxRequestSize

openAIHelper

protected GPTInternalOpenAIHelper openAIHelper

Constructor Detail

GPTDictationServiceImpl

public GPTDictationServiceImpl()

Method Detail

activate

protected void activate(GPTDictationServiceImpl.GPTDictationServiceConfig config)
                 throws URISyntaxException

Throws:

URISyntaxException

deactivate

protected void deactivate()
                   throws IOException

Throws:

IOException

modified

protected void modified(GPTDictationServiceImpl.GPTDictationServiceConfig config)
                 throws IOException,
                        URISyntaxException

Throws:

IOException

URISyntaxException

isAvailable

public boolean isAvailable(@Nullable
                           GPTConfiguration configuration)

Description copied from interface: GPTDictationService

Whether the service is enabled and properly configured.

Specified by:

isAvailable in interface GPTDictationService

transcribe

public String transcribe(@Nonnull
                         InputStream audioStream,
                         @Nonnull
                         String contentType,
                         @Nullable
                         String language,
                         @Nullable
                         GPTConfiguration configuration,
                         @Nullable
                         String prompt)

Description copied from interface: GPTDictationService

Transcribes the input audio to text.

Specified by:

transcribe in interface GPTDictationService

Parameters:

audioStream - the audio stream to transcribe, will be closed

contentType - the content type of the audio, e.g. "audio/mpeg" for mp3, "audio/wav" for wav

language - the language code to use, e.g. "en" for English, or null for automatic detection

configuration - the configuration to use, or null for the default configuration

prompt - an optional prompt to give the AI some context, e.g. previous sentences