Speech to Text

The IBM Watson Speech to Text service uses speech recognition capabilities to convert Arabic, English, Spanish, French, Brazilian Portuguese, Japanese, and Mandarin speech into text.


Transcribe Audio

Use your microphone (compatible only with Google Chrome and Mozilla Firefox). Upload pre-recorded audio (WAV for uncompressed audio, FLAC or OPUS) file formats. Drag and drop recorded audio onto the page, or use the audio samples provided. The returned result includes the recognized text, word alternatives (aka confusion networks), and spotted keywords. You may choose to spot your keywords by entering them (separated by commas) in the text box.

Would you like to help make this service better?

Allow Watson to learn from this session
Opt out
Keywords Spotted
    Word Alternatives Hide alternate words
    Word Alternatives will appear shortly after audio transcription is started