Speaking up: Automatic Text to Speech voiceovers for Substack posts?
What I've learned about availability of the automatic TTS feature in Substack, and some options for moving forward.
Audio voiceover can make texts more accessible to subscribers with deafness or hearing impairments, or with more flexibility to listen than to read.
However, creating audio voiceovers can be challenging for writers.
If you’re already a follower, you know that I’ve been looking into AI-based and non-AI based options for adding audio voiceovers to my posts. (My recent articles on this topic are linked in the endnotes.1)
But what if text-to-speech capabilities were automatic and free in Substack?
Thanks to a Note reply by
on Friday, I discovered that Substack might already have this exact feature!
✅The Good News
I did some searching on Friday and found a link which confirmed2 that Substack does have an automatic text-to-speech (TTS) feature. 😮 And I found two posts, one dating to June 2022 and one from April 2024, which explain the supported listening features.3
❌The Bad News
Unfortunately, according to the Substack chat support app (transcript in Appendix), there are 4 key limitations on the feature:
TTS is supposedly enabled on “most” publications to be “automatically available” - but not for all publications.
Writers cannot configure or control whether automatic TTS is enabled for their publications.
The criteria for rollout of TTS aren’t public, and there is no timeframe for it being enabled for all.
Automatic TTS is an app-only feature (not available on website).
The only way to know if a publication has TTS enabled is to look for a PLAY icon on one of its posts. (Audio voiceover manually added by the writer will override automatic TTS.)
Although it’s been nearly 2 years since the rollout of the feature was announced, it looks like automatic TTS has not been enabled on any of my new publications (all created in 2024)🙁
❓ For writers here: Does YOUR publication have automated TTS enabled?
❓ For readers here: Do any publications YOU subscribe to have automated TTS enabled?
Alternatives
The chatbot politely suggested that I could manually add voiceovers to my posts. I’m already experimenting with that, and it’s certainly different than auto-TTS from a personal touch perspective.
However, not all writers have access to (or can afford) good quality recording equipment, a quiet & non-echoing room to record without interruption, etc. And it takes time, work, and practice to get a good quality voiceover recording without ums, ahs, coughs, background noises, etc.
This is one reason I’ve been looking into AI-based voice cloning tools - hoping they will minimize the effort required to generate audio recordings, at minimal cost. But none of those tools are free for reasonable volumes (minutes of recording per month).
Wouldn’t it be amazing if all of us didn’t have to shell out that money, invest that time to make audio recordings, and become or hire audio engineers (or editing tools) to achieve good-enough quality to benefit our readers & listeners?
GOAL: make Substack newsletter posts more accessible
to READERS
with deafness or hearing impairment, or with more flexibility to listen than to read,
and to WRITERS
with limited time, money, or audio production skills to make high quality recordings.
What’s Next?
I found a link to a “beta signup form” for automatic TTS enablement in the June 2022 article. I’ve now submitted my publication URLs, and will let you all know what I hear. (If you sign up too, let us know if you get in!)
In the meantime, I’m continuing my experiments with manual recordings and AI-based generation of audio voiceovers - results to be reported in future articles in
.I’ve shared below the links I’ve collected on podcasting and adding audio voiceovers, in case they are useful to others who are also just getting started.4
APPENDIX: Transcript of my May 10, 2024 Substack support chat on automatic TTS
Me: I saw "If text-to-speech is available on your publication, a voiceover in the post will override the TTS." on a Support page. How do I enable text-to-speech on one of my publications?
SS: Hey Karen!
Text-to-speech (TTS) is a feature that's automatically available on most Substack publications, but not all. If you don't see this option on your publication, it might be enabled at a later time as it's not something that can be manually turned on by the author.
If you're looking to add a personal touch, you can add a voiceover to your posts instead. This will override the TTS feature where available. Would you like a quick rundown on how to add a voiceover?
Me: No, I already did a few voiceovers today. I'm looking for the automatic TTS feature. Where would I see if it's enabled on mine? Why is it not available on all?
SS: Got it, Karen. If text-to-speech is available on your publication, you would typically see a play icon on your posts in the Substack app. This is where readers can tap to listen to your posts using the TTS feature.
As for why it's not available on all publications, it's a feature that's being rolled out gradually. It may become available to your publication automatically in the future, but there isn't a manual setting for it at the moment. Keep an eye out for updates, as we're always working on expanding our features.
If you need any more info or assistance, feel free to ask!
Me: is the TTS feature only available in the Substack app, not on the website?
SS: Yes, that's correct! The text-to-speech feature is currently only available in the Substack app. It's not an option on the website. So, if your readers want to listen to your posts, they'll need to use the app on their mobile devices. (read more)
Me: Is TTS planned to be added to the website?
SS: As of now, there's no specific information available about whether text-to-speech will be added to the website in the future. We're always working on improving and adding new features, so it's possible that this could change. If you'd like, I can log this as a feature request for our team to consider! Would that be helpful for you?
Me: Yes, please do
SS: Great, I've logged your request for text-to-speech functionality on the website for our team to consider. We appreciate your input as it helps us improve Substack. If there's anything else you need help with, just let me know!
Me: When will TTS be available on ALL publications and not just "most"?
SS: Currently, there isn't a specific timeline for when text-to-speech will be available on all publications. It's being rolled out gradually, and availability may increase over time. We understand that having TTS on all publications would be beneficial, and we're working towards improving the feature's reach. Keep an eye on updates from us, as we'll share news when more publications have access to TTS. If you have any other questions or need further assistance, feel free to ask!
Me: Saying "availability MAY increase over time" implies that it might not increase over time. Are there criteria for which publications get TTS, e.g. a minimum number of posts?
SS: I apologize for any confusion. The criteria for enabling text-to-speech on publications have not been explicitly detailed, so I can't provide specific requirements such as a minimum number of posts. It seems to be a feature that's rolled out at our discretion and may be influenced by various factors that aren't publicly disclosed.
If you're concerned about the availability of TTS for your publications, I can certainly pass along your interest to our team. It's feedback like yours that helps us prioritize what features to develop and expand. If there's anything else I can assist you with, just let me know!
Me: Yes, please pass along that it would be very desirable for accessibility to have TTS on all publications, not just "most", and to have it in the website as well as the app. Thank you.
SS: Absolutely, Karen. Your feedback about the importance of text-to-speech for accessibility on all publications and on the website has been noted. We truly value your input as it helps us prioritize features that enhance the experience for all users. Thank you for bringing this to our attention. If you have any more suggestions or need further assistance, just reach out!
“How do I add a voiceover to my Substack post?”, Substack support article mentioning TTS being enabled in some publications; last updated around May 2, 2024
“New ways to listen to your favorite writers”, on.substack.com, June 28, 2022
“How do I listen to a Substack post?”, Substack support article, last updated in April 2024
Tips on voiceovers and audio recordings:
“What is voiceover? A complete guide for beginners”, Descript, Vivian Tejeda, Sept. 10, 2023
“How to do a voice over like a pro: The complete guide”, TechSmith, Ryan Knott, Aug. 2, 2023
“Everything you need to know about creating high-quality voiceover”, Well Said Labs, Martin Ramirez, Jan. 12, 2022
“13 more tips to help you add narration like the pros”, Articulate, June 9, 2009
“The radical act of being cool with your voice”, Samara Bay, Feb 21, 2024
“How to create a podcast (bonus issue)”, Edward Nevraumont, May 20, 2022 - covers audio recording and editing as well as podcast-specific points like interviewing
“The 8 best audio recording software picks for any device and budget”, Descript, Brandon Copple, November 6, 2023
I googled (as I have before) to figure out how to enable text-to-speech because they just included it again in their On Substack post. I arrived at your post. It's good to know there is nothing I can manually do to enable it. My posts still do not have it and yet the majority of those that I read have it enabled. I wish I could know why I am in the minority.
It's been 2.5 weeks; no word yet on being allowed into the beta program for automated voiceovers. So far I've manually recorded 4 post voiceovers for a total of about 32 minutes. The recording process wasn't as painful and repetitive as I feared. The audio time runs about 50-100% longer than the estimated reading time. I'm not sure yet where to see in the post metrics whether anyone has played the audio.