Speech Analysis

Convert speech to text and analyse it using a text analysis tool

Converts speech in audio or video format to text. Optionally, send the converted speech text to one of the text analysis tools. Currently supported file formats for this demo (more file formats may be supported by the providers): AWS Transcribe: mp3 | mp4 | wav and Azure Speech to Text: wav.

  1. Select from one of the example media or upload your own.
  2. The analysis occurs and may take some time.
  3. Transcription results from available providers will be displayed below (Note, max 150 words shown which is approx 1 minute of speech).
  4. Optionally, send text to one of our text analysis demos by clicking the Send to Text Analysis button.

Pick an example:

More information about this demo

Speech to text is well known and can work in a wide variety of languages. Speech to text can be used to transcribe anything which involves speech such as lectures and talks. Speech to text can also be used to improve accessibility for deaf people, for example.

Things to consider

Speech is a powerful tool and can be used to convey information (text tagging demo), sentiment (sentiment analysis demo) and tone of voice (tone analysis demo). Speech analysis could be used as a tool to help connect students to information, gather feedback from students (both information and sentiment/tone), as a tool to help student helpline calls determine sentiment or tone of voice, etc.

Speech to text works well for clearly annunciated speech in neutral accents but struggles with thick accents and quick speech, similar in respects to humans.