Unity 在 Microsoft Azure Text-To-Speech 处理输入时冻结 2 秒

Question

我正在使用带有 Unity 的 Microsoft Azure Text To Speech。它的工作原理是当按下按钮从文本输入中产生语音时，整个应用程序冻结大约 2 秒，然后输出声音，游戏恢复正常。我认为此冻结是由于 Azure 正在处理 TTS？下面是代码。

public void ButtonClick()
    {
        // Creates an instance of a speech config with specified subscription key and service region.
        // Replace with your own subscription key and service region (e.g., "westus").
        var config = SpeechConfig.FromSubscription("[redacted]", "westus");

        // Creates a speech synthesizer.
        // Make sure to dispose the synthesizer after use!
        using (var synthsizer = new SpeechSynthesizer(config, null))
        {
            lock (threadLocker)
            {
                waitingForSpeak = true;
            }

            // Starts speech synthesis, and returns after a single utterance is synthesized.
            var result = synthsizer.SpeakTextAsync(inputField.text).Result;

            // Checks result.
            string newMessage = string.Empty;
            if (result.Reason == ResultReason.SynthesizingAudioCompleted)
            {
                // Since native playback is not yet supported on Unity yet (currently only supported on Windows/Linux Desktop),
                // use the Unity API to play audio here as a short term solution.
                // Native playback support will be added in the future release.
                var sampleCount = result.AudioData.Length / 2;
                var audioData = new float[sampleCount];
                for (var i = 0; i < sampleCount; ++i)
                {
                    audioData[i] = (short)(result.AudioData[i * 2 + 1] << 8 | result.AudioData[i * 2]) / 32768.0F;
                }

                // The default output audio format is 16K 16bit mono
                var audioClip = AudioClip.Create("SynthesizedAudio", sampleCount, 1, 16000, false);
                audioClip.SetData(audioData, 0);
                audioSource.clip = audioClip;
                audioSource.Play();

                newMessage = "Speech synthesis succeeded!";
            }
            else if (result.Reason == ResultReason.Canceled)
            {
                var cancellation = SpeechSynthesisCancellationDetails.FromResult(result);
                newMessage = $"CANCELED:\nReason=[{cancellation.Reason}]\nErrorDetails=[{cancellation.ErrorDetails}]\nDid you update the subscription info?";
            }

            lock (threadLocker)
            {
                message = newMessage;
                waitingForSpeak = false;
            }
        }
    }

    void Start()
    {
        if (inputField == null)
        {
            message = "inputField property is null! Assign a UI InputField element to it.";
            UnityEngine.Debug.LogError(message);
        }
        else if (speakButton == null)
        {
            message = "speakButton property is null! Assign a UI Button to it.";
            UnityEngine.Debug.LogError(message);
        }
        else
        {
            // Continue with normal initialization, Text, InputField and Button objects are present.
            inputField.text = "Enter text you wish spoken here.";
            message = "Click button to synthesize speech";
            speakButton.onClick.AddListener(ButtonClick);
            //ButtonClick();
        }
    }

我希望 TTS 在按下 TTS 按钮时不会冻结整个应用程序，因此当按下 TTS 按钮时应用程序是可用的。任何帮助将不胜感激。

Answer 1

当您执行synthesizer.SpeakTextAsync(inputField.text).Result; 时，它会阻塞直到任务完成。相反，尝试调用Task<SpeechSynthesisResult> task = synthesizer.SpeakTextAsync(inputField.text);，然后设置一个协程，直到task.IsCompleted() 为真，然后执行代码中的其余过程

这里有一个部分（未经测试的）解决方案可以帮助您入门。我将变量从 synthsizer 更改为 synthesizer，并删除了所有锁定，因为协程在主线程上按顺序发生，因此不需要锁定：

public void ButtonClick()
{
    if (waitingForSpeak) return;

    waitingForSpeak = true;

    // Creates an instance of a speech config with specified subscription
    // key and service region.
    // Replace with your own subscription key and service region (e.g., "westus").
    SpeechConfig config = SpeechConfig.FromSubscription("[redacted]", "westus");

    // Creates a speech synthesizer.
    // Make sure to dispose the synthesizer after use!       
    SpeechSynthesizer synthesizer = new SpeechSynthesizer(config, null));

    // Starts speech synthesis, and returns after a single utterance is synthesized.
    Task<SpeechSynthesisResult> task = synthesizer.SpeakTextAsync(inputField.text);

    StartCoroutine(CheckSynthesizer(task, config, synthesizer));
}

private IEnumerator CheckSynthesizer(Task<SpeechSynthesisResult> task, 
        SpeechConfig config, 
        SpeechSynthesizer synthesizer)
{
    yield return new WaitUntil(() => task.IsCompleted());

    var result = task.Result;

    // Checks result.
    string newMessage = string.Empty;
    if (result.Reason == ResultReason.SynthesizingAudioCompleted)
    {
        // Since native playback is not yet supported on Unity yet (currently
        // only supported on Windows/Linux Desktop),
        // use the Unity API to play audio here as a short term solution.
        // Native playback support will be added in the future release.
        var sampleCount = result.AudioData.Length / 2;
        var audioData = new float[sampleCount];
        for (var i = 0; i < sampleCount; ++i)
        {
            audioData[i] = (short)(result.AudioData[i * 2 + 1] << 8 
                    | result.AudioData[i * 2]) / 32768.0F;
        }

        // The default output audio format is 16K 16bit mono
        var audioClip = AudioClip.Create("SynthesizedAudio", sampleCount, 
                1, 16000, false);
        audioClip.SetData(audioData, 0);
        audioSource.clip = audioClip;
        audioSource.Play();

        newMessage = "Speech synthesis succeeded!";
    }
    else if (result.Reason == ResultReason.Canceled)
    {
        var cancellation = SpeechSynthesisCancellationDetails.FromResult(result);
        newMessage = $"CANCELED:\nReason=[{cancellation.Reason}]\n"+
                     $"ErrorDetails=[{cancellation.ErrorDetails}]\n"+"
                     "Did you update the subscription info?";
    }

    message = newMessage;
    waitingForSpeak = false;
    synthesizer.Dispose();
}

void Start()
{
    if (inputField == null)
    {
        message = "inputField property is null! Assign a UI InputField element to it.";
        UnityEngine.Debug.LogError(message);
    }
    else if (speakButton == null)
    {
        message = "speakButton property is null! Assign a UI Button to it.";
        UnityEngine.Debug.LogError(message);
    }
    else
    {
        // Continue with normal initialization, Text, InputField and Button 
        // objects are present.
        inputField.text = "Enter text you wish spoken here.";
        message = "Click button to synthesize speech";
        speakButton.onClick.AddListener(ButtonClick);
        //ButtonClick();
    }
}

---

作为对 cme​​ts 的回应，这里是另一种方法的开始，您可以尝试将数据复制包装在一个任务中并在该任务完成之前产生：

private IEnumerator CheckSynthesizer(Task<SpeechSynthesisResult> task, 
        SpeechConfig config, 
        SpeechSynthesizer synthesizer)
{
    yield return new WaitUntil(() => task.IsCompleted());

    var result = task.Result;

    // Checks result.
    string newMessage = string.Empty;
    if (result.Reason == ResultReason.SynthesizingAudioCompleted)
    {
        // Since native playback is not yet supported on Unity yet (currently
        // only supported on Windows/Linux Desktop),
        // use the Unity API to play audio here as a short term solution.
        // Native playback support will be added in the future release.

        Task copyTask = Task.Factory.StartNew( () => 
        {
            var sampleCount = result.AudioData.Length / 2;
            var audioData = new float[sampleCount];
            for (var i = 0; i < sampleCount; ++i)
            {
                audioData[i] = (short)(result.AudioData[i * 2 + 1] << 8 
                        | result.AudioData[i * 2]) / 32768.0F;
            }

            // The default output audio format is 16K 16bit mono
            var audioClip = AudioClip.Create("SynthesizedAudio", sampleCount, 
                    1, 16000, false);
            audioClip.SetData(audioData, 0);
            audioSource.clip = audioClip;
            audioSource.Play();
        });

        yield return new WaitUntil(() => copyTask.IsCompleted());

        newMessage = "Speech synthesis succeeded!";
    }
    else if (result.Reason == ResultReason.Canceled)
    {
        var cancellation = SpeechSynthesisCancellationDetails.FromResult(result);
        newMessage = $"CANCELED:\nReason=[{cancellation.Reason}]\n"+
                     $"ErrorDetails=[{cancellation.ErrorDetails}]\n"+"
                     "Did you update the subscription info?";
    }

    message = newMessage;
    waitingForSpeak = false;
    synthesizer.Dispose();
}