How to Convert Speech to Text using Google Speech-to-Text REST API

June 9, 2023

How to Convert Speech to Text using Google Speech-to-Text REST API

There are two possible ways to write the code.

(1) Upload the Audio File to Google Cloud Storage

First write code to upload the audio file to Google Cloud Storage.
For example, see:

Next, write the code to make the Speech-to-Text REST API call. You can use Chilkat’s online tool at https://tools.chilkat.io/curlHttp to convert the following CURL statement to code using Chilkat in your chosen programming language.

Note that the uri references the audio file previously uploaded to Google Cloud Storage.

curl -H "Content-Type: application/json" \
     -H "Authorization: Bearer [YOUR_ACCESS_TOKEN]" \
     --data '{
       "config": {
         "encoding":"FLAC",
         "sampleRateHertz": 16000,
         "languageCode": "en-US"
       },
       "audio": {
         "uri":"gs://your-bucket/audio.flac"
       }
     }' \
     "https://speech.googleapis.com/v1/speech:recognize?key=[YOUR_API_KEY]"

(2) Base64 Encode the Contents of the Audio File and Send Base64 in the REST API POST

You can alternatively base64 encode the contents of the audio file and include the Base64 directly in the REST API POST.
To base64 encode, simply load the file into a Chilkat BinData object, then call BinData.GetEncoded(“base64”) to get the data as base64.

curl -H "Content-Type: application/json" \
     -H "Authorization: Bearer [YOUR_ACCESS_TOKEN]" \
     --data '{
       "config": {
         "encoding":"FLAC",
         "sampleRateHertz": 16000,
         "languageCode": "en-US"
       },
       "audio": {
         "content": "[BASE64_ENCODED_AUDIO_DATA]"
       }
     }' \
     "https://speech.googleapis.com/v1/speech:recognize?key=[YOUR_API_KEY]"

Again, use Chilkat’s online tool at https://tools.chilkat.io/curlHttp to convert the above CURL statement to source code.

admin