Speech to Text : Specifically for captchas
This is no ordinary speech to text API, it is specifically designed to crack audio captchas
If you use AWS Transcribe, or Google Cloud Speech to Text on a captcha audio, then you will have poor results, because it’s a general-purpose speech to text API, designed to transribe video, narrated text, and phone calls. This API is different, it is designed to quickly and accurately solve the short, distorted, random letter and number assortment found in captcha audio.
Audio can be provided as a URL or Base64 encoded data
Standard alphabet and NATO alphabet supported (Alpha, Bravo, Charlie …)
Returns on average in 5 seconds.
You can always call the API multiple times, most websites don’t count failed attempts.
Read more about this new API at https://www.audiocaptcha.com/
Create an account on Rapid API, and get an API key for this API. Once done, you can try out the API for free, by setting a HTTP header “x-rapidapi-key” to your API Key, then posting to the following URL:
“url” : “https://pepite.s3.eu-west-1.amazonaws.com/65UHC.wav”,
“base64” : “”,
“useNato” : false
With the above JSON – Obviously, the Wave file URL here is a demo, but it was extracted from a real captcha.
Otherwise, you can provide the audio in base 64 format in the base64 field and omit the URL element.
If the audio is in the NATO alphabet (Alpha, Bravo, Charlie …) then you change useNato to true, otherwise, it’s assumed to be (A,B,C,D …)