DeepSpeed

by Microsoft

Don’t want to install DeepSpeed? Drop the file here.

The first 20 minutes of any recording are transcribed free: a recording up to 20 minutes transcribes in full, a longer one is previewed, and a pack from $6 (500 minutes) transcribes the rest and opens every export, and credits never expire.

Distributed Whisper / Conformer training at scale.

Linux  ·  free

TL;DR

Best for fine-tuning Whisper-large on a single 80GB GPU via ZeRO. Pricing: free.

Category
Open source
License
Apache-2.0
Stars
★ 42.9k
Last push
2026-08-07
Pricing
free
Platforms
Linux

What it is

Microsoft's distributed-training framework. Apache-2.0.

Best for: Fine-tuning Whisper-large on a single 80GB GPU via ZeRO.
Watch out for: Linux + NVIDIA only.

Install / use

pip install deepspeed

What people actually do with DeepSpeed-style transcription

The tool is the means. These are the jobs — each one priced at published rates, each one wired up on its own page.

Wire it into Make.comGoogle DriveEmergentNotionYouTubeAmazon S3

Features

Speaker diarizationNo
Word-level timestampsYes
Streaming / real-timeNo
Languages supported99
HIPAA eligibleNo

Links

GitHub repo ↗

DeepSpeed vs Whipscribe

FeatureDeepSpeedWhipscribe
CategoryOpen sourceTranscription APIs
Pricingfree$6–$24 one-time packs (credits never expire) · first 20 minutes free
Speaker identificationNoNamed in the summary, key quotes and chapters
Word timestampsYesYes
StreamingNoNo
Languages9999
PlatformsLinuxWeb, API, MCP

Alternatives to DeepSpeed

Frequently asked about DeepSpeed

Is DeepSpeed free?

Yes. DeepSpeed is free and open source.

Does DeepSpeed work on Mac, Windows and Linux?

DeepSpeed runs on Linux. It is not listed for other platforms; check the vendor for anything beyond Linux.

How do I install DeepSpeed?

With pip: `pip install deepspeed`. You need Python and, for most audio tools, ffmpeg on your PATH first.

How many languages does DeepSpeed support?

DeepSpeed lists 99 languages. That is the Whisper-family multilingual set, so quality varies by language — English and the large European languages are strongest.

Does DeepSpeed identify different speakers?

No. DeepSpeed does not label speakers; a conversation comes back as one continuous text. If you need speaker labels, that is a separate tool or a different service.

Does DeepSpeed give word-level timestamps?

Yes — DeepSpeed produces timing per word, which is what subtitle cues and clip boundaries need.

Can DeepSpeed transcribe live audio?

No — DeepSpeed works on finished files, not a live stream.

Is DeepSpeed HIPAA compliant?

DeepSpeed does not claim HIPAA compliance. For PHI, run it on infrastructure you control or choose a service that will sign a BAA.

What are the limitations of DeepSpeed?

Linux + NVIDIA only.

Who is DeepSpeed best for?

Fine-tuning Whisper-large on a single 80GB GPU via ZeRO.

Does DeepSpeed run offline?

DeepSpeed runs on your own machine — it is open source, so nothing leaves the computer unless you configure it to.

Can I use DeepSpeed in the browser?

No — DeepSpeed is a Linux application, not a web app. You install it rather than sign in to it.

Where is the DeepSpeed source code?

On GitHub at github.com/microsoft/DeepSpeed. It is maintained by Microsoft.

What kind of tool is DeepSpeed?

In this directory DeepSpeed is filed under training, distributed as a open-source tool.

What are the alternatives to DeepSpeed?

There is a side-by-side page at /tools/diff-svc-deps-deepspeed-alternatives comparing DeepSpeed with the closest tools in the same category on price, platform and features.

Whipscribe is a hosted transcription service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.

Transcribe your recordings

Explore

Have the recording already?

Skip the setup and send the file or the link: see how to transcribe an MP3 of any length, a podcast episode as text from its link and the SRT file generator.

Looking at paid options?

Before you choose a transcription tool or service: see an honest roundup of the best transcription services, the cheapest way to transcribe an hour of audio and transcription services from $6.