• The Javascript SDK has had a major update to include all of our latest API features. Check out the changes here: https://github.com/aflorithmic/audiostack-js/blob/HEAD/CHANGELOG.md
• Easier sound design and voices browsing:
  • The Dictionary has been updated, and now allows you to search for voices;
  • You can browse sound designs directly in SonicSell.
• Platform updates:
  • You can now see private (cloned) voices in the new Platform;
  • You can now resubscribe to a previous plan; and,
  • Workflows page has been given a makeover.

You can now find our guide to generating a complete audio advert using AI, in the API Recipes sections of the docs. Check it out 🚀

You can also now annotate speech with timestamps, to generate transcripts or identify when certain words are spoken. Find out more ⏱️

Platform Launch

We spent months redesigning our enterprise-grade audio workflows.

Try it out here https://platform.audiostack.ai/ 💯

What can I do with this

• Turn text into beautiful audio ads in seconds with SonicSell

• Explore our diverse sound library

• Single Sign on - now sign in effortlessly with your Google account
• Explore our Voice Library with over 1300 voices!

• Dictionary - custom pronunciations for all AI voices
• Enterprise-grade admin and analytics functionality to control who has access to your data.
• Try it out here https://platform.audiostack.ai/ 💯

Lots more coming so please feel free to send us feedback to support[at]audiostack[dot]ai we're happy to partner with such great partners.

• We've updated our documentation to make it easier to find what you need to get started, and to clarify how the API is structured.
• We also added documentation on available SSML Tags: https://docs.audiostack.ai/docs/ssml-tags
• Voice Cloner documentation has also been added: https://docs.audiostack.ai/docs/ai-voice-cloner
• AdLocal documentation added: https://docs.audiostack.ai/docs/adlocal

Remove Padding from Speech

Many of the voice providers in our library add padding to the end of speech files by default, so that sentences don't end too suddenly. There are some situations where users have reported needing to remove this padding so that they can specify a more precise, consistent amount of padding across different voices and providers. This is now possible using /speech/tts/remove_silence.

M4A Encoding

🔊 AudioStack now supports M4A encoding. Simply specify "m4a" as the preset to use this feature.

Javascript SDK Quickstart

We added a quick start guide to get Javascript developers up and running using our SDK. You can find it here.

If you have any feedback about our documentation or would like to request docs on specific workflows, please get in touch at support@audiostack.ai.

 450 NEW VOICES ADDED 🎉

• We added 450+ new voices to the Voice Library
• Including some new voices such as Nivedita which is a Malayalam 🇮🇳 voice.
• We also added some new voices from 🇰🇷, 🇳🇴, 🇧🇷 and many more!
• You can check them out here https://library.audiostack.ai/
• Let us know at support@audiostack.ai what your favourite voice is for the chance to win a prize! 📥

Platform Improvements [beta]

We added AdLocal to the Beta in https://platform.audiostack.ai

This enables you to run localised national radio or podcast campaigns.

A screenshot of some of the options 💯

Bug Fix - Voice Intelligence layer

• Our customers reported a 🐛 where the dictionary was incorrectly updated, this turned out to be an edge case involving some ~being passed into one of our systems. Well done to the engineering team for hunting this one down, this was a hard one! 🎩
• We are constantly working on bug fixes and improvements to our systems.

Improved Incident Response Systems ⏰

• We invested heavily this week in improving our internal alerting systems - this is to further enhance our reliability. The improvements were rolled out this week 💯

 Voice Intelligence Layer Quality Enhancements

• We added 2500 tests to our continuous integration system, enhancing the quality of our product and the customer experience.

A Couple More Layered Sound Templates... [beta]

• We've made sound layering available for the following templates:ambient_electronica, eighties_chill
• These now have intense, soft and default intensity options. For more information about the layered sound template beta, see our previous ChangeLog post.

Sound Template Intensity (Beta feature)

• New feature allows you to specify how intense you would like a specific sound template to sound, changing the "timbre" of the sound design.
• For example, a more intense sound template might include heavier percussion and more complex melodies. A soft version of the template might sound more sparse, with just melodic instruments and chords, for example.
• To use this feature, you need to specify which of the "soundLayers" to use in the mix: choose from soft, intense, or default.
• This feature can currently be used with two supported sound templates: chromatic_jazz and friendly_electronica.

import audiostack
import os

audiostack.api_key = "APIKEY"
audiostack.api_base = "https://v2.api.audio"

print(audiostack.Production.Mix.list_presets())

scriptText2 = """
<as:section name="intro" soundSegment="intro">
Insert your script for the intro section here
</as:section>

<as:section name="main" soundSegment="main">
And this is the main section!
</as:section>

<as:section name="outro" soundSegment="outro">
Here is your outro.
</as:section>
"""


script = audiostack.Content.Script.create(scriptText=scriptText2)
print("response from creating script", script.response)
scriptId = script.scriptId

# create one tts resource
tts = audiostack.Speech.TTS.create(scriptItem=script, voice="bryer")
print(tts.speechId)

sound_template = "chromatic_jazz"
sound_layer = "intense"
mix = audiostack.Production.Mix.create(speechItem=tts, soundTemplate=sound_template, masteringPreset="musicenhanced", soundLayer=sound_layer)# sectionProperties={"main" : {"endAt" : 120}})
print(mix)


encoded = audiostack.Delivery.Encoder.encode_mix(productionItem=mix, preset="wav")
encoded.download(fileName=f"{sound_template}_{sound_layer}")

👍

We love to receive feedback about beta features

If you have any thoughts on this feature, please get in touch at bamber@audiostack.ai.

We'll be launching a more extended version of this feature which supports lots more sound templates soon.

Mastering Improvement

• We added MP3 support for media files in mastering - this means you can upload MP3 media files and include them in your final asset

🚧

Don't forget to update the SDK to access new features

In Python, you can do this by running the command pip3 install -U audiostack

Bug fixes and code quality enhancements

• Bug fix for joining speech files back together when there was an error, customer facing bug
• We improved the code quality in one of our core services improving your customer experience
• New SDK version https://pypi.org/project/audiostack/

 Platform update

• New features in Platform. Save audio assets locally, so that your session is saved, improving the customer experience.

New mastering preset 🎉

• We added a "voice only" preset (mastering engine feature) 💯
• By using this preset, mastering will only applied to the generated voice and not the sound design beneath it. This is perfect for cases where you've already applied mastering to the sound template (or received a finished advert that you want to add an audio tag to).

Code Example:

mix = audiostack.Production.Mix.create(
    speechItem=tts,
    soundTemplate="pulsing_guitar",
    masteringPreset="voiceonly",
)

Docs

• We added a new guide, with code example, to support workflows where users need to add an audio "tag" to the end of an existing audio advert: <https://docs.audiostack.ai/docs/advanced-timing-example-add-a-tag-to-the-end-of-an-advert>

Terms of use and Legal

• Updated our terms of use - https://audiostack.ai/terms-of-use to make the licence terms of our sound templates clearer and easier to understand.

Sound templates

• New sound templates live. We added 14 more templates, covering a range of genres and styles.
• You can see a full list of our templates here: https://docs.audiostack.ai/docs/sound-templates-and-sound-designs

Speech / Predict

• We've added a new endpoint which draws on 4 million data points to predict the duration of speech from a given script and list of voice names 🤓
• This will allow users to determine whether a voice is suitable, or whether their script is the correct length for their project, without needing to generate the audio itself, saving time and credits! ⏰
• Find out more and check out a code example in our guide, in the API reference or in the video below 👇

📢 We've added documentation and code examples on how to generate an audio "tag" and add it to the end of an existing audio advert. Find out more here.