> For the complete documentation index, see [llms.txt](https://autoedit.gitbook.io/user-manual/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://autoedit.gitbook.io/user-manual/master-1.0.13/setup-stt-apis.md).

# Setup: STT APIs

There are three options for speech to text APIs that you can use with this system.

Check them out individually for extra setup instruction.

1. [IBM Watson STT](/user-manual/master-1.0.13/setup-stt-apis/setup-stt-apis-ibm.md)
2. [Speechmatics](/user-manual/master-1.0.13/setup-stt-apis/setup-stt-apis-speechmatics.md)
3. [AssemblyAI](/user-manual/master-1.0.13/setup-stt-apis/setup-stt-apis-assemblyai.md)
4. [Gentle STT](/user-manual/master-1.0.13/setup-stt-apis/setup-stt-apis-gentle.md) (Open Source, needs a separate app for setup)
5. Pocketsphinx (Open Source, integrate inside of autoEdit, no extra setup needed)
6. [Rev - Transcriptions service](/user-manual/master-1.0.13/setup-stt-apis/setup-stt-apis-rev.md)

## Overview <a href="#overview" id="overview"></a>

### IBM Watson STT Service <a href="#ibm-watson-stt-service" id="ibm-watson-stt-service"></a>

#### Pros:

* 16 hours a month included in service. 0.02 cent a minute after that [see pricing](https://console.ng.bluemix.net/catalog/services/speech-to-text).
* Speed. In autoEdit, always takes 5 minutes to transcribe any duration of audio or video.
* Generally pretty accurate (my opinion, judge for yourself)
* Supports a number of languages [see here](https://console.ng.bluemix.net/catalog/services/speech-to-text). Including distinction between British and American English.

#### Cons:

* Need to provide card details for pay as you go fee.
* is in the cloud so no offline support.

[Check out for extra setup instructions for IBM](/user-manual/master-1.0.13/setup-stt-apis/setup-stt-apis-ibm.md).

### Speechmatics STT Service <a href="#ibm-watson-stt-service" id="ibm-watson-stt-service"></a>

* 1 hour free credit with new account&#x20;
* Easy to setup credentials&#x20;
* Generally pretty accurate (my opinion, judge for yourself)
* [28 languages, see full list](https://www.speechmatics.com/language-support/) Including support for "accent agnostic global english".

### AssemblyAI STT Service <a href="#ibm-watson-stt-service" id="ibm-watson-stt-service"></a>

* &#x20;Free tier: 5 hours free per month
* Competitive pricing at $0.0003 per second
* Easy to setup credentials&#x20;
* Generally very accurate (my opinion, judge for yourself)
* For now only support for English but more coming soon

### Gentle Open Source STT <a href="#gentle-open-source-stt" id="gentle-open-source-stt"></a>

#### Pros:

* Free as in free speech as well as in free beer.
* Working locally on your machine. No internet connection needed because of that, good for sensitive material.
* Open source [github repo](https://lowerquality.com/gentle) and [I made a node module to work with the API](https://github.com/OpenNewsLabs/gentle_stt_node) .

#### Cons:

* Not as accurate as IBM one (in my opinion, but decide for yourself).
* Only support US english STT.
* In autoEdit, at the moment not as fast as IBM one, takes a little longer then the length of the media. (eg 27 min takes 30 min to transcribe).

[Check out for extra setup instructions for Gentle](https://github.com/pietrop/autoEdit_2_user_manual/tree/d5c8cea5ec4e2a1cee11515e8a838d832407badc/setup-stt-apis-gentle.md).

### Pocketsphinx Open Source STT <a href="#pocketsphinx-open-source-stt" id="pocketsphinx-open-source-stt"></a>

#### Pros:

* Free as in free speech as well as in free beer.
* working locally on your machine. no internet connection needed because of that, good for sensitive material.
* [Open source module in autoEdit ](https://github.com/OpenNewsLabs/autoEdit_2)originally extracted from [Videogrep](https://github.com/antiboredom/videogrep) project.

  .

#### Cons:

* Not as accurate as IBM or Gentle.
* Only support US english STT.
* in autoEdit, at the moment not as fast as IBM one, takes a little longer then the length of the media. (eg 27 min takes 30 min to transcribe).

Pocketsphinx does not require extra setup to use.

### Rev Transcription service <a href="#rev-transcription-service" id="rev-transcription-service"></a>

#### Pros:

* [Transcription service from real humans](https://www.rev.com/transcription)
* In theory more accurate then automated transcriptions
* word level timestamps&#x20;

#### Cons:

* Price more expensive then automated transcriptions
* Turnaround slower then automated transcriptions.

{% hint style="info" %}
Sign up to the [mailing list](http://eepurl.com/cMzwSX), follow on [twitter](http://twitter.com/autoEdit2) and/or [facebook](https://www.facebook.com/autoEdit.io/) to keep up to date with the latest releases. Say hi at [pietro@autoEdit.io](mailto:pietro@autoEdit.io?Subject=Hello), always curious to hear what autoEdit is helping you with.
{% endhint %}

{% hint style="info" %}
[autoEdit.io](http://www.autoEdit.io) it's free and open source. Free as in free speech as well as in free beer. [Help support the autoEdit project to keep it that way](https://donorbox.org/c9762eef-0e08-468e-90cb-2d00643697f8?recurring=true). Support will go towards fixing bugs, adding features, provide support for users etc...
{% endhint %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://autoedit.gitbook.io/user-manual/master-1.0.13/setup-stt-apis.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
