SSML是否会增加Google文本到语音配额使用情况？
我有一个使用TTS API的小型独立项目，并且想利用SSML在某些情况下改变文本的阅读方式，但试图避免在4m chars/month的配额下进行避免收取费用（可能是一堆堆的钱。
我想知道所有标记是否都符合配额，但找不到有关此的任何详细信息。
示例：
（21个字符）
This is a test. Boop!
与
（141个字符）
<speak>This is a <prosody rate="50%">test.</prosody> <audio src="https://some website/boop.wav">Boop!</audio></speak>
在检查控制台时我实际上看不到字符使用情况的任何内容（总是说“此期间没有数据”），在给定月份中，我从未真正知道我有多么接近该限制。因此，如果对此添加支持将使配额使用率增加约5倍，我想知道。
每分钟请求的配额： https://i.sstatic.net/bpikf.png br> 所有请求中字符的计数： https：// https://i.sstatic.net/jfuke.png
另外，如果有人碰巧知道那件char量的事情，我将有兴趣听到。
I have a little indie project that uses the TTS API, and want to make use of SSML to change the way the text is read in some circumstances, but trying to stay under the 4M chars/month quota to avoid getting charged (potentially a bunch of) money.
I'm wondering if all the markup counts towards the quota, but can't find any detailed information about this.
Example:
(21 chars)
This is a test. Boop!
vs.
(141 chars)
<speak>This is a <prosody rate="50%">test.</prosody> <audio src="https://some website/boop.wav">Boop!</audio></speak>
As I can't actually see anything about character usage when I check the console, (always says "No data for this period") I never actually know in a given month how close I am to exceeding that limit. So if adding support for this would increase quota usage by a factor of ~5x, I'd like to know.
Quota of requests per minute: https://i.sstatic.net/BpIkF.png
Count of chars in all requests: https://i.sstatic.net/jfuKe.png
Also, if anybody happens to know what's up with that char-count thing, I'd be interested in hearing.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
根据此文本转语音定价 GCP 文档，语音合成标记语言(SSML) 标签也包含在计费字符数中，因此您可能需要相应增加配额。有关最佳实践和提示以及在文本转语音中使用 SSML，您可以参考此 文档。
As per this Text-to-Speech pricing GCP Documentation, Speech Synthesis Markup Language (SSML) tags are also included in the character count for billing, so you may have to increase your quota accordingly. For best practices and tips and using SSML in Text-to-Speech, you can refer to this documentation.