Speech-to-Text & Text-to-Speech for Next.js: what 76 makers' products chose
Among 76 makers' products built with Next.js that we know the Speech-to-Text & Text-to-Speech of, ElevenLabs leads: 44 of them (58%). Next come Deepgram (21) and Cartesia (12).
Real-time or batch, and which direction. Streaming voice features need low latency above everything, while transcribing recordings is about accuracy and price per hour. Most providers lead in one direction — speech-to-text or text-to-speech — so many apps end up using two.
What makers on Next.js chose
| Option | Next.js products | All makers | 2026 launches |
|---|---|---|---|
| 44 (58%) | 56% | 63% | |
| 21 (28%) | 27% | 27% | |
| 12 (16%) | 12% | 17% | |
| 11 (14%) | 11% | 7% | |
| 7 (9%) | 9% | 10% | |
| 3 (4%) | 4% | 7% | |
| 1 (1%) | 0% | 0% |
Live makers' products with a source for both: that they're built with Next.js, and what they use for Speech-to-Text & Text-to-Speech. A product using two options counts in both; not market share. "All makers" is the option's share among every maker's product that made this decision. "2026 launches" is its share among the 30 Next.js products launched that year: a tool that came first has products from every year since, so its total runs ahead.
Which to pick, by situation
- How the voice sounds is part of the product — narration, characters, content
ElevenLabs 44 here
Fish Audio 3 here - You're building live conversation where every millisecond of delay shows
Cartesia 12 here
Deepgram 21 here - Users upload recordings to be transcribed and analysed
AssemblyAI 11 here
Deepgram 21 here - Your product needs recordings and transcripts from Zoom, Meet or Teams calls
Recall.ai 7 here - Voice is a side feature and one vendor matters more than fine control
OpenAI API 1 here
ElevenLabs
Voice generation44 Next.js products here (58%), against 56% of all makers' products.
In Lovable, ReplitBest forNatural, expressive voices and voice cloning for narration, content and characters.
Trade-offCredits run out fast on long audio, and cloning needs consent from the voice owner.
“Create natural AI voices instantly in any language The best AI voice service out there right now. Fast, reliable, and ahead of the game. Pretty much a no-brainer”
“We use ElevenLabs to create natural AI voices in any language. ElevenLabs also provide the best voice quality, including clones of real-world personalities”
Speech-to-Text & Text-to-SpeechMaker says so +1 · source ↗“Create natural AI voices instantly in any language The best AI voice service out there right now. Fast, reliable, and ahead of the game. Pretty much a no-brainer”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“I'm a big fan of ElevanLabs TTS solutions.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“For launch video SFX”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Creating natural, human-like voice interactions powered by AI.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Low latency and great responses”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“We've used ElebenLabs voices on our platform and it has been the smoothest experience – thank you!”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Human-quality voices and a reliable API service.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“We're open-source and BYOK, so we integrate every TTS provider - and users still pick ElevenLabs most. Latency is competitive now, but the emotional range on longer calls is what keeps prospects on the line. For outbound voice agents, voice quality is the We're open-source and BYOK, so we integrate every TTS provider - and users still pick ElevenLabs most. Latency is competitive now, but the emotional range on longer calls is what keeps prospects on the line. For outbound voice agents, voice quality is the difference between a conversation and a hang-up.difference between a conversation and a hang-up.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗
Deepgram
Transcription21 Next.js products here (28%), against 27% of all makers' products.
Best forStreaming speech-to-text with low latency, for live calls and voice interfaces.
Trade-offIts text-to-speech is built for voice agents, with a set of prebuilt voices and no voice cloning.
“We love using Deepgram for real time transcription in Martin's advanced voice mode and phone call interactions.”
“We use Deepgram for STT. It is still one of the best for transcription of RealTime AI”
Speech-to-Text & Text-to-SpeechMaker says so +1 · source ↗“Great STT models. Fastest streaming transcription we tested.”
Speech-to-Text & Text-to-SpeechMaker says so +1 · source ↗“Deepgram is the current voice platform that transcribes audio in the mac app.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Delivering fast, accurate, and scalable speech recognition capabilities.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“We use Deepgram to transcribe live AI-driven training calls, where SDRs and sales reps practice conversations with an AI agent impersonating real prospects. The fast, accurate transcription is essential — it enables our coaching system to deliver instant feedback and correction right after each mock call. This real-time loop helps reps improve faster, and we’re thankful to the Deepgram team for enabling it.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“We love using Deepgram for real time transcription in Martin's advanced voice mode and phone call interactions.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Transcribes videos super fast”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“For enabling the fastest Speech to text Latency”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“The fastest speech to text there is”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“The long-time leader in realtime speech-to-text models and APIs.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗
Cartesia
Voice generation12 Next.js products here (16%), against 12% of all makers' products.
Best forText-to-speech with very low time to first audio, for live conversation.
Trade-offBuilt for real-time use, with fewer creator tools such as dubbing and sound effects.
“Cartesia is the best TTS model in terms of latency. Many of our clients use Cartesia to solve their latency issues”
“Cartesia is the best TTS model in terms of latency. Many of our clients use Cartesia to solve their latency issues”
Speech-to-Text & Text-to-SpeechMaker says so +1 · source ↗“One of our voice providers”
Speech-to-Text & Text-to-SpeechMaker says so +1 · source ↗“Pretty natural-sounding voice :)”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Cartesia Sonic powers our voice preservation technology, enabling us to maintain the natural tone and personality of each speaker while translating in real-time. The low latency and human-like quality are essential for natural conversations.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“We love Cartesia for their super-fast and hyper-realistic voices!”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Low latency, natural, and easy to use.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Great for Text-to-Speech. Sonic 2 is the best model in our opinion in terms of voice quality, speed, and cost.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Cartesia stood out for its ultra-low latency voice synthesis and natural-sounding speech. It helped us create more human-like real-time conversations while keeping the developer experience simple and flexible.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“The best voices at the moment!”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗
AssemblyAI
Transcription11 Next.js products here (14%), against 11% of all makers' products.
Best forAccurate transcription of recordings with speaker labels and extras like summaries and topic detection.
Trade-offFocused on speech-to-text; you need another vendor for speech output.
“AssemblyAI gives us consistently accurate speech-to-text with fast turnaround and a simple API, which performs more reliably than other transcription services we tested.”
“Helping us turn live conversations into the context Convo needs in real time.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“One of the best and seamless APIs I've every used. Super easy to get started and use. Very accurate too.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“AssemblyAI gives us consistently accurate speech-to-text with fast turnaround and a simple API, which performs more reliably than other transcription services we tested.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“AssemblyAI made it easy to build production-ready voice workflows with accurate transcription, clean APIs, and great documentation. Their startup support helped us experiment and iterate quickly.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“AssemblyAI provides us with accurate transcripts that enhance our subtitles, editing processes, and more.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Easily the best ASR model out there. I have tested all the popular options for speech recognition, Assembly AI consistently outputs the best transcription with high accuracy.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Assembly's transcription API is fantastic.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“When selecting a speech-to-text & audio intelligence provider for NeuraVid.io, we compared several options, including Deepgram, Rev AI, AWS Transcribe, and Whisper-based APIs. We ultimately chose AssemblyAI because of the following key advantages: superior transcription quality, multilingual & the best speaker diariaztion that we tested”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗
Recall.ai
Meeting bots7 Next.js products here (9%), against 9% of all makers' products.
Best forGetting recordings, transcripts and participant data from Zoom, Google Meet, Microsoft Teams and other calls through one bot API.
Trade-offBilled per recording hour on top of any transcription, and the bot is visible to everyone in the call.
“Our fellow YC alumni accelerated our product development and helped us optimized our integrations with Zoom and MS Teams.”
“Simplicity and ease of use”
Speech-to-Text & Text-to-SpeechMaker says so +1 · source ↗“They are an awesome partner - great support.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Meeting bot as a service - alternative - Meeting BaaS, recall.ai is really competitive.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Our fellow YC alumni accelerated our product development and helped us optimized our integrations with Zoom and MS Teams.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Using them for our bot solution and they are great.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“Recall.ai has a reliable API solution that allows us to offer meeting features to our customers.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗
Fish Audio
Voice generation3 Next.js products here (4%), against 4% of all makers' products.
Best forText-to-speech and quick voice cloning with a large library of community voices in 30+ languages.
Trade-offCommunity voices vary in quality and rights; its open-weight models are under the Fish Audio Research License, so check its terms before self-hosting commercially.
“I needed audio for about 5,000 Japanese dialogue lines. Most TTS makes Japanese sound like a train announcement. Fish Audio was the first one where my two characters sounded like actual people, and the price didn't scare me off.”
“We chose Fish Audio for four reasons: reliability, quality, speed, and the team. They’ve been consistently stable at scale, the output quality is among the best we tested, latency is fast enough for a real-time experience, and the team moves quickly and collaborates closely with us.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗“I needed audio for about 5,000 Japanese dialogue lines. Most TTS makes Japanese sound like a train announcement. Fish Audio was the first one where my two characters sounded like actual people, and the price didn't scare me off.”
Speech-to-Text & Text-to-SpeechMaker says so · source ↗Also chosen, and not seen yet
Fewer Next.js products here: OpenAI API (1).
Real options for Speech-to-Text & Text-to-Speech not seen among these Next.js products yet: Whisper, Kokoro.
Getting Speech-to-Text & Text-to-Speech right
Decide whether audio is live (calls, voice UI) or recorded (uploads, meetings); the right provider differs. Test with your own audio — accents, background noise and domain terms change accuracy more than benchmarks show. For TTS, listen to full paragraphs, not single sentences, before choosing a voice.
- Choosing a speech-to-text provider on clean benchmark audio, then discovering it struggles with your users' accents, jargon and background noise.
- Cloning a voice without the speaker's documented consent, which breaks provider terms and, in many places, the law.
- Sending whole recordings to a streaming API or live audio to a batch one, paying for latency you don't need or waiting on latency you can't accept.
The rest of the Next.js stack
What makers' Next.js products chose for their other decisions, the most common pick for each:
- Hosting: Vercel 1,078 of 1,312
- Transactional Email: Resend 493 of 830
- Database: Supabase 171 of 382
- Payments: Stripe 288 of 359
- LLM API: OpenAI API 222 of 301
- Authentication: Clerk 176 of 211
- Web Analytics: Google Analytics 105 of 197
- Product Analytics: PostHog 123 of 168
Questions
What do most Next.js apps use for Speech-to-Text & Text-to-Speech?
Among 76 makers' products built with Next.js that we know the Speech-to-Text & Text-to-Speech of, ElevenLabs leads: 44 of them (58%). Next come Deepgram (21) and Cartesia (12).
How this is counted: a product is on Next.js when a live-product source says so (the maker, its website, a customer story or a subprocessor list), and the same for its Speech-to-Text & Text-to-Speech. How evidence is collected.