API Comparison Table

Heading 1

Heading 2

Heading 3

Heading 4

Heading 5
Heading 6

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.

Block quote

Ordered list

  1. Item 1
  2. Item 2
  3. Item 3

Unordered list

Text link

Bold text

Emphasis

Superscript

Subscript

Pricing
Get started
Get started

Read more

Speech-To-Text

Speech-to-text for AI medical scribes: Why clinical vocabulary breaks generic STT

TL;DR: Generic STT engines fail in clinical environments because language model probability overrides correct acoustic detection of medical terms, substituting phonetically plausible but clinically wrong candidates silently. The result corrupts drug names, dosages, and diagnoses before the LLM ever sees them. Before selecting an STT engine for a medical scribe, verify four things: whether vocabulary biasing works at inference time without fine-tuning, whether async diarization accurately separates clinician and patient audio, whether the model holds up on noisy consultation recordings rather than clean read-speech, and whether the vendor's data training policy covers PHI by default on your plan.

Speech-To-Text

Migrating from self-hosted Whisper to a managed speech-to-text API

TL;DR: Self-hosting Whisper's true cost rarely sits in the model weights. GPU idle time, VRAM leaks under parallel load, and the engineering hours spent maintaining CUDA dependencies and diarization pipelines are where the bill compounds. For teams processing under roughly 3,000 hours per month, assuming 20% of one US FTE at $150K loaded annual cost, a managed API is cheaper, though the break-even shifts materially against your actual labor cost. Above that threshold, the decision depends on your DevOps overhead and whether audio accuracy on real-world recordings matters for downstream systems like CRM sync and coaching scores.

Speech-To-Text

Migrating from AssemblyAI to Gladia: A step-by-step switching guide

TL;DR: Switching from AssemblyAI requires four concrete changes: update one auth header, remap batch endpoints, adjust the JSON response schema, and resample audio for WebSocket connections. Multiple customers independently report completing these in under a day with a rollback abstraction layer in place. The bigger structural difference is cost model: a production stack with diarization, sentiment, entities, and summarization runs $0.30/hr on AssemblyAI's Universal-2 tier because each feature is metered separately, versus a bundled base rate. This guide covers the exact parameter mappings, payload diffs, WebSocket reconfiguration, and a zero-downtime cutover strategy.

March 2023 Roadmap its Speech-to-Text API: Speaker Diarization, Word-Level Timestamps and more

Published on Jun 2, 2023
March 2023 Roadmap its Speech-to-Text API: Speaker Diarization, Word-Level Timestamps and more

A glimpse into Gladia's roadmap for its Speech-to-Text API, starting with speaker diarization. We’re incredibly excited to be building our Audio Intelligence product in a community-led way, delivering a holistic final product adapted to the many needs and use cases brought to our attention.

Following Gladia’s Speech-to-Text AI alpha release two weeks ago, we’ve received dozens of new feature requests from the alpha users, to make our core real-time audio transcription API even more exciting and versatile.

We heard you and are happy to announce that the API is growing more robust by the minute and is now available with more capabilities — on top of its blazing speed and top-tier output quality.

We’re incredibly excited to be building our Audio Intelligence product in a community-led way, delivering a holistic final product adapted to the many needs and use cases brought to our attention.

Here’s what we have in store already

Speech-to-Text (STT) Transcription

Setting a new standard for the industry, our STT API is build on OpenAI’s Whisper and can transcribe audio in 10s/h at 3.52%WER. Tested and approved by thousands of alpha users across a range of use cases (e.g. call center, virtual meetings, YouTube videos, podcasts).

Speech-to-Text Translation

Upload your file, select an output language of your choice, and enjoy the final translated transcript free of errors. Currently available in 99 languages, and counting. If your language is not supported yet, drop us a message in this Twitter thread.

Transcription from YouTube URL

Drop a video URL and enjoy a highly accurate output file (.srt or JSON) that can be used as an alternative to YouTube’s auto-captions to improve the viewer’s experience on your channel. Transcription as subtitles file (.srt) will become available shortly too.

And here is the list of new most anticipated features we’re planning to release in March.

Speaker Diarization

You will now be able to automatically identify and recognize all speakers mixed in a single audio or video stream, including when multiple languages are used.

Word-Level Timestamps

A feature enabling Gladia users to produce a highly accurate JSON transcript with time stamps at every word.

Live-Streaming Transcription

We’re adding the ability to transcribe speech in real-time, using your microphone.

We’re preparing a series of deep dives on some of these new features to showcase how our tech works behind the scenes. Stay tuned!

As always, feel free to test the API and give us your feedback  Discord. We truly love iterating with the community.

Contact us

280
Your request has been registered
A problem occurred while submitting the form.

Read more