Verbi is a speech-to-text API

It listens to a recording or a live feed and hands back words your code can use: who said what, when, and how surely.

• Send a file. One HTTP call, one complete transcript back.
• Let Verbi listen live. Over a WebSocket, each utterance arrives about a second after the pause that ends it.

Verbi bills only for the speech it hears. Silence is free.

And it hears more than words: music, singing and everyday sounds — a knock, a cough, a barking dog, laughter, typing — each with the moment it started and ended, in files and live streams alike. No need to ask, and no extra charge.

Verbi also tells speakers apart, times every word with confidence and spots the language — all of that in 25 European languages — naturally, by default. Ask, and it writes SRT/VTT captions and spells your rare names right from a glossary.

Speed

• File uploads: a single file at about 50 times real time; files in parallel add up to about 300 times in total (over 1,000 with voice detection off).
• Live stream: an utterance's text typically arrives under half a second after it ends.

Getting started

Nothing to set up on your side — no GPU, no model downloads, no scaling, no on-call. Integrate using the recipes below, or, easier still, ask your agent.

# no email, no card — mint a key and transcribe
KEY=$(curl -s -XPOST "https://verbi.services/v1/signup" | python3 -c 'import sys,json;print(json.load(sys.stdin)["api_key"])')

# transcribe a file — or several in parallel; any audio or video format,
# one JSON reply each: text, words with timestamps, speakers, music
# (a public sample here — for your own file, use --data-binary @your-file.mp3)
curl -s "https://verbi.services/live-sample.wav" \
  | curl -s -XPOST "https://verbi.services/v1/transcribe" -H "Authorization: Bearer $KEY" --data-binary @-

# or a live source — a radio channel, an RTSP camera, a microphone — over
# a WebSocket, the connection the internet is happy to hold open for hours:
# the transcript streams back utterance by utterance, one JSON message per line
# (swap the -i address for your own source: a stream URL, an RTSP camera, a file)
ffmpeg -v fatal -i "https://npr-ice.streamguys1.com/live.mp3" -ac 1 -c:a libopus -b:a 24k -f ogg - \
  | websocat -b --text-prefix $'\n' "wss://verbi.services/v1/transcribe" -H "Authorization: Bearer $KEY"

# more recipes and agent-optimized docs:
# https://verbi.services/llms.txt

Limits

• 2 transcriptions at a time without a verified email, 10 with one, 25 once paid.
• Verbi reads no faster than it transcribes: a faster upload simply waits.

What Verbi does not do, so you find out now rather than in week three: no word-by-word realtime ASR (the unit is a complete utterance, about a second behind the speaker); no lyrics — sung vocals are kept out of the transcript and the bill, and under mode=vad_diar_off, which transcribes everything, lyrics come out poorly; no SOC 2 or standing DPA yet; speaker labels are measured in-house, not independently audited.

That's all there is to it

The whole product is one page: /llms.txt — every endpoint, every limit, every known weakness. That page plus a couple of hands-on experiments on the free tier is all you need to make a decision.

And there's no retention policy to read, because nothing is ever retained — not the audio, not the transcript, not a voiceprint — less to worry about, and part of why the whole product fits on one page.

Write to contact@verbi.services with anything — feedback, a bug, a gap in the docs, a feature you want. And if a limit, the throughput or the price is in your way, say so: committed-volume plans raise or remove limits and come with volume discounts, and complex custom integrations are possible too.

For machines: /llms.txt · OpenAPI
For people: API docs · ReDoc