Pocket FM gives its writers an AI tool to transform narratives, write cliffhangers, and more

Audio series platform maker based in India Pocket FM strives to be the Netflix of Audio. That is, the company is planning to match its audio series with hundreds of episodes to the taste of its users. To make that work, it must quickly release content – something that it now turns to AI to help with.

The startup supported by Lightspeed gives its writers an AI tool set that can do things such as better endings for an episode suggest or make the story more attractive. The hope is that the tools will speed up the story -writing process.

Pocket FM already uses some AI tools such as Elflabs to generate voices for audio series. It also tested AI tools for writing and adaptation aid internally.

Rohan Nayak, founder of Pocket FM, said it rolls out the AI tools for all writers, so it takes them less time to end their episodes.

The writing tool, dubbed Second pilotCan be used to help every writer make a story.

Copilot can transform storage -based writing into dialogue -based writing for a specific segment. It can also do “Beat analysis” to shape writing in a way to make it more attractive for an audio series of a certain genre. The tool also has simple writing functions in Chatbot style such as “Korten”, “Expand”, and the possibility to generate text through a prompt.

To build Copilot, the company investigated thousands of hours of data points to understand what uses more with a certain storyline in a specific genre.

Based on that, it added the functions for writing suggestions that were designed to increase the conflict between characters and recommend endings for an episode to make it more exciting. AI can also suggest tags for background effects that can be used to produce the audio.

The tool can automatically generate bios of characters, generate their relationships and summarize plot points of various episodes, allowing makers to refer to these details while writing.

Copilot also has an assessment tool, which checks for plot points, grammar and quality feedback through comments about an episode.

Under the hood, Pocket FM trains smaller models to maintain the context for a story for character arches and relationships, together with narrative consistency. In addition, the use of signals from users, the startup AI starts to add more drama to the story.

International expansion and localization plans

In addition to the arrival of the AI tools, Pocket FM has launched adjustment tools for different markets that not only translate the text from one language to another, but also change names and sentences that are more suitable for the culture of that region.

The company debuted this tool for the first time as part of the Copilot Suite in Germany earlier this year to convert stories from other regions after allegedly Struggle to involve users in the European country last year.

Nayak said that the company saw great results of this test, with a constant increase in the monthly in-app income, which exceeded $ 700,000 in June.

“When we started expanding to new regions, it took us 12-18 months to make sense in that market. You must have at least 1,000 hours of content to acquire and scale the market. Now we can do this in less than three months,” Nayak said.

The tool increased the productivity of the writer by a maximum of 50% for the German market in terms of Showoutput. Moreover, the tool helped the company to make more flawless concepts of the shows that resulted in a higher user retention for audio series.

In the US, series made with the help of this new AI tools now contribute 10% of the playing time. In addition, these shows have generated $ 7 million in income in the last 12 months, while the production costs are reduced 2-3 times.

Building technology to scale content generation

As a result of the internal use of different AI functions internally, Pocket FM was able to scale the content quickly. The startup said it is launching nearly 1,000 pilots per month. And only the enormous number of content means that a few of them become hits.

But the audio show is only a part. The company is already working on tools to convert stories into comic strips with his Pocket Toon platform. Moreover, Nayak said that video is a possible format that the company could also explore. The startup, which has collected more than $ 196 million in financing, is also experimenting with a micro -drama app.

By next year, Pocket FM wants to release its own Singular Large Language Model (LLM), which will be based on data collected from the shows and contains various tools, such as writing aid, adoption, dramatization and preservation of story context. The company said that when it is over to its own LLM, it does not have to train a ton of small models for individual functions.

AI’s potential disadvantages

Aciting AI has had its side effects.

Pocket FM has already fired people who were employees or contractors over multiple rounds in the last 12 months. There have also been reports of seeing writers reduced efficiency over time. And the company is confronted Legal cases in California on employment and payslips.

“Like most industries guided by the content, we work with a diverse network of writers, speech artists and production partners on a project basis, so that resources are adjusted to every market. AI has had a minimal influence on our Core Creative Community; instead, the new roads opened to expand the reach and the output, a company reference signed.”

There are also questions about quality. The company measures quality through the retention numbers of a show.

The basic argument is that the new AI tools acts as a writers room, even for solo makers, so that they can quickly produce more content. Moreover, on the basis of the figures, writers can quickly edit the story with the help of AI. However, these tools can very well cause “ai-slop” or low quality, AI-generated content-in the platform and can influence the recommendations of a user, making it difficult to discover good stories.

Pocket FM argues that stories with a solid structure will become popularity, even though AI helps them.

The company noted that every piece of content is assessed by its AI-driven fashion frame to guarantee quality and originality. It also claims that his AI moderation controls on things like duplication, copyright issues, content health and other quality measures before you approve audio to go live. Each show receives an equal push and user involvement ultimately determines the ranking of a show.

Another concern is that writers can become overly dependent on AI over time.

In Germany, AI writes more content than people per show for selected titles. With Pocket FM’s plans to roll out more AI tools, the amount of AI written content can increase. And with that the expectation of expressing more shows can also increase. Unless the acceptance of users also rises rapidly, the average return can fall.

The company does not directly relate to TechCrunch questions about the return, but said that the AI tools can accelerate the work of a writer and help them edit an episode based on figures and feedback from the public. That is that they can make targeted improvements instead of doing a complete rewriting.

“In this way faster content does not necessarily dilute the quality or relevance; it simply shifts the role of the writer to editing, refining and refining and managing more productive output,” a spokesperson said in a statement.

We always want to evolve, and by giving some insight into your perspective and feedback in WAN and our coverage and events, you can help us! Fill up This investigation To let us know how we do it and get the chance to win a prize in exchange for that!