How to Convert the Audio File to Flac for Google Speech Recognition

Shane Steven How to

#1 Why I need to convert audio file to Flac format.

Google recommends using FLAC or LINEAR16, the lossless codec to transmit audio and you can get the results you expect from the API. Just as it clarifies in its document:

For Optimal Results:

Use a lossless codec to record and transmit audio. FLAC or LINEAR16 is recommended.

If Possible, avoid:

Using mp3, mp4, m4a, mu-law, a-law or other lossy codecs during recording or transmission may reduce accuracy. If your audio is already in an encoding not supported by the API, transcode it to lossless FLAC or LINEAR16. If your application must use a lossy codec to conserve bandwidth, we recommend the AMR_WB, OGG_OPUS or SPEEX_WITH_HEADER_BYTE codecs, in that preferred order.

Therefore, in order to get higher accuracy and best results, you need to convert your audio file to FLAC, especially, only 1 channel(mono). Luckily, you can perform this process by using some free and high-efficient online tools. (Do not worry, we will integrate this feature in our program soon)

#2 How to convert audio files(like .mp3) to Flac fast and free.

Here, I recommend you using online-convert.com to process audio file to .flac file.

Click here to the online tool.

convert audio files to flac

It is quite simple and fast to perform the process.

  • Upload your audio file.
  • Choose the MONO channel.
  • Start conversion and download.

Then, you can drag and drop your FLAC audio file to EasySpeech2Text and then convert the speech to text.

Leave a Reply