Yes, IBM Watson’s Speech to Text API can handle MP3 audio files, as long as they meet its supported audio encoding requirements (e.g., MP3 with proper bitrate and sampling rate). The API also supports formats like WAV, FLAC, and Ogg, but you’ll run into issues with things like WMA, AAC without proper container support, or proprietary codec formats — those are among the more common unsupported formats.

If you just need a quick way to convert MP3 speech to text without API setup, keys, or audio re-encoding, I’ve built a speech to text browser extension that works directly in your browser. You just drag & drop your MP3, and it uses modern automatic speech recognition (Whisper by OpenAI) to give you a transcript.

You can check it out here: https://chromewebstore.google.com/detail/speech-to-text/jolafoahioipbnbjpcfjfgfiililnoih