ONE MCP CONNECTION · ASR
Audio transcription & speech to text through MCP
Bring a spoken idea back into text. Transcribe an audio recording for editing, reference or a draft, with optional timing segments for subtitles.
What you can do
- Transcribe an audio file from an accessible HTTPS URL.
- Specify the language when it helps recognition.
- Request timing segments for subtitle work.
- Copy text or export SRT when your host displays the interactive transcript widget.
Try this brief
Example prompt · not a claim that this output has already been generated
Transcribe https://lavis.app/site/examples/creative-flow.mp3 in English. Include timestamps, and keep the wording faithful to the recording.
Available tools: asr_transcribe
REAL OUTPUT · OCTOBER 9, 2026
From audio back to words.
This sample was rendered using Lavis’s Fish Audio speech adapter (s2.1-pro-free) and transcribed by Fish Audio ASR. It is an operator-created demonstration, not a customer recording.
Actual transcription
A small idea can become something beautiful. Create a picture, give your story a voice, and keep your creative flow with Lavis Studio.
The speech sample uses the same provider adapter as Lavis. It does not demonstrate a particular host’s MCP connection or widget. This published sample is retained as website content; user-generated media links expire after 7 days.
Models and pricing
| Model | Starting rate |
|---|---|
| Fish Audio ASRfish-asr | 2 credits / request |
1 credit ≈ US$0.01. Settings and reference inputs can affect the final quote. Current account pricing takes precedence.
Before you start
Transcripts may contain recognition errors. Review names, numbers and punctuation before publication. Timing segments are not a guarantee of word-perfect alignment.
Connect with remote MCP over Streamable HTTP and OAuth. Google sign-in creates your Lavis account. Interactive previews require MCP Apps support; download files you want to keep before their 7-day expiry.