GenshinTextToSpeech: 基于原神语音 API 的 Unity 文本转语音工具

public class GenshinTextToSpeech : TTS {

#region 参数定义
/// <summary>
/// Genshin配置项
/// </summary>
/// <summary>
/// 朗读的角色
/// </summary>
[Header('朗读声音设置')]
public string speaker = '胡桃';
/// <summary>
/// 音频格式
/// </summary>
[Header('音频格式设置')]
public string format = 'wav';
/// <summary>
/// 音频长度
/// </summary>
[Header('音频长度设置')]
public float length = 1f;
/// <summary>
/// 噪声
/// </summary>
[Header('噪声设置')]
public float noise = 0.5f;
/// <summary>
/// 噪声权重
/// </summary>
[Header('噪声权重设置')]
public float noisew = 0.9f;
/// <summary>
/// 声调比例
/// </summary>
[Header('声调比例设置')]
public float sdp_ratio = 0.2f;

#endregion

private void Awake()
{
    m_PostURL = 'https://genshinvoice.top/api'; 
}

/// <summary>
/// 语音合成
/// </summary>
/// <param name="_msg"></param>
/// <param name="_callback"></param>
public override void Speak(string _msg, Action<AudioClip> _callback)
{
    StartCoroutine(GetVoice(_msg, _callback));
}

/// <summary>
/// 语音合成，返回合成文本
/// </summary>
/// <param name="_msg"></param>
/// <param name="_callback"></param>
public override void Speak(string _msg, Action<AudioClip, string> _callback)
{
    StartCoroutine(GetVoice(_msg, _callback));
}

/// <summary>
/// restful api语音合成
/// </summary>
/// <param name="_msg"></param>
/// <param name="_callback"></param>
/// <returns></returns>
private IEnumerator GetVoice(string _msg, Action<AudioClip> _callback)
{
    stopwatch.Restart();
    //发送报文
    string textToSpeechRequestBody = GenerateTextToSpeech(speaker, _msg, format, length, noise, noisew, sdp_ratio);
    Debug.Log('传参的值是：' + textToSpeechRequestBody);
    using (UnityWebRequest speechRequest = new UnityWebRequest(m_PostURL, 'POST'))
    {
        byte[] data = System.Text.Encoding.UTF8.GetBytes(textToSpeechRequestBody);
        speechRequest.uploadHandler = (UploadHandler)new UploadHandlerRaw(data);
        speechRequest.downloadHandler = (DownloadHandler)new DownloadHandlerAudioClip(speechRequest.uri, AudioType.WAV);

        yield return speechRequest.SendWebRequest();

        if (speechRequest.responseCode == 200)
        {
            AudioClip audioClip = DownloadHandlerAudioClip.GetContent(speechRequest);
            _callback(audioClip);
        }
        else
        {
            Debug.LogError('语音合成失败: ' + speechRequest.error);
        }
    }

    stopwatch.Stop();
    Debug.Log('Genshin语音合成耗时：' + stopwatch.Elapsed.TotalSeconds);
}

/// <summary>
///  restful api语音合成，返回合成文本
/// </summary>
/// <param name="_msg"></param>
/// <param name="_callback"></param>
/// <returns></returns>
private IEnumerator GetVoice(string _msg, Action<AudioClip, string> _callback)
{
    stopwatch.Restart();
    //发送报文
    string textToSpeechRequestBody = GenerateTextToSpeech(speaker, _msg, format, length, noise, noisew, sdp_ratio);
    Debug.Log('传参的值是：' + textToSpeechRequestBody);
    using (UnityWebRequest speechRequest = new UnityWebRequest(m_PostURL, 'POST'))
    {
        byte[] data = System.Text.Encoding.UTF8.GetBytes(textToSpeechRequestBody);
        speechRequest.uploadHandler = (UploadHandler)new UploadHandlerRaw(data);
        speechRequest.downloadHandler = (DownloadHandler)new DownloadHandlerAudioClip(speechRequest.uri, AudioType.WAV);

        yield return speechRequest.SendWebRequest();

        if (speechRequest.responseCode == 200)
        {
            AudioClip audioClip = DownloadHandlerAudioClip.GetContent(speechRequest);
            _callback(audioClip, _msg);
        }
        else
        {
            Debug.Log('传参的值是：' + ?);
            Debug.LogError('语音合成失败: ' + speechRequest.error);
        }
    }

    stopwatch.Stop();
    Debug.Log('Genshin语音合成耗时：' + stopwatch.Elapsed.TotalSeconds);
}

}