Speech Assistant
Available nowLanguageTranscription, synthesis and transcript intelligence.
Speech-to-text and text-to-speech on the shared runtime, plus transcript summarisation and language detection. Audio is sent to the runtime for the request and is never stored by this platform.
Speech to text
Upload an audio file and receive a written transcript.
Accepted formats: WAV, MP3, MP4/M4A and WebM. Your audio is not stored by this platform.
Text to speech
Enter your text and generate audio you can play or download.
Transcript intelligence
Paste transcript text to get a summary or action items.
Hi — I'm your speech assistant. Transcription, synthesis and transcript intelligence.
Tell me what you're working on and I'll take it from there. If I need a couple of details to do it properly, I'll ask.
For example:
- Transcribe a recorded call or voice note
- Turn a transcript into a decision summary
- Generate a spoken version of written content
How to use this product
What it does
When to use it
Expected inputs
- · Transcript (required)
- · File upload (optional) — audio/wav, audio/mpeg, audio/mp4, audio/webm, audio/x-m4a, up to 20 MB
Expected outputs
- · Transcript summary: formatted text — Summarise a transcript into decisions and actions.
- · Language detection: structured JSON — Detect the language of a transcript.
- · Speech-to-text
- · Text-to-speech
- · Audio summaries
- · Language detection
- · Transcript export
Technical details
Use cases
Transcribe a recorded call or voice note
Turn a transcript into a decision summary
Generate a spoken version of written content
Detect the language of a recording before routing it
