speech.h

/* Copyright 2017 ID R&D Inc. All Rights Reserved. */
 
#pragma once
#include <voicesdk/core/config.h>
#include "speech_summary.h"
 
#include <vector>
#include <memory>
#include <iostream>
 
namespace voicesdk {
 
    class VOICE_SDK_API SpeechEndpointDetector {
    public:
 
        using Ptr = std::shared_ptr<SpeechEndpointDetector>;
 
        static SpeechEndpointDetector::Ptr create(
                size_t minSpeechLengthMs,
                size_t maxSilenceLengthMs,
                size_t sampleRate);
 
        virtual void reset() = 0;
 
        virtual void addSamples(const uint8_t* bytes, size_t bytesNum) = 0;
 
        virtual void addSamples(const float* samples, size_t samplesNum) = 0;
 
        virtual void addSamples(const int16_t* samples, size_t samplesNum) = 0;
 
        virtual bool isSpeechEnded() = 0;
 
        virtual ~SpeechEndpointDetector() = default;
    };
 
    class VOICE_SDK_API SpeechEndpointDetectorOpus {
    public:
 
        using Ptr = std::shared_ptr<SpeechEndpointDetectorOpus>;
 
        static SpeechEndpointDetectorOpus::Ptr create(
                size_t minSpeechLengthMs,
                size_t maxSilenceLengthMs,
                size_t sampleRate);
 
        virtual void reset() = 0;
 
        virtual void addPacket(const uint8_t *bytes, size_t bytesNum) = 0;
 
        virtual bool isSpeechEnded() = 0;
 
        virtual ~SpeechEndpointDetectorOpus() = default;
    };
}