Alternatives
Products that do what DeepSpeech based automated transcription service does
We have been building a DeepSpeech model with our data for the past year and we have recently hit 95% accuracy on the LibriSpeech dataset. That puts us close to the published results for DeepSpeech 2. However our dataset is conversational audio and we do much better with our own internal dataset compared to PaddlePaddle. Here's a blog post on the method we followed to build our models. https://scribie.com/blog/2018/03/continual-learning-for-spee... We have been using this internally in our service and it saves a ton of time and effort during the typing stage. It…
- 1
- 2
- 3
The most accurate streaming speech model for voice agents.
Mar 2026
- 4
- 5IC
A friend and I wrote a book on how to build and train Deep Learning models in Go. We wanted it to be a useful reference for deep learning basics for Go programmers. Deep Learning is slowly seeping into everything we use every day and we thought it would be great if more people could do it in Go. The book is available here and on Amazon as well. https://www.packtpub.com/big-data-and-business-intelligence/hands-deep-learning-go We would appreciate any feedback and we're always looking to improve.
2019
- 6IB
Hey HN! I built Transcrib.ee to help me generate transcripts for lectures on YouTube with no captions, especially multilingual lectures. I was constantly frustrated with videos that had no transcripts or inaccurate ones, so I built a tool and decided to share it with everyone. This tool uses Groq (an AI inference engine that's different from Grok and offers very fast processing) and OpenAI's Whisper model (really accurate in multilingual) to quickly transcribe any YouTube video, regardless of the language. It's been a game-changer for my studies! To make it even faster, I created a Chrome…
2024 · transcrib.ee
- 7MM
Hi HN! We (Thomas and Stéphan, hello!) recently released Model2Vec, a Python library for distilling any sentence transformer into a small set of static embeddings. This makes inference with such a model up to 500x faster, and reduces model size by a factor of 15 (7.5M params or 15/30MB on disk, depending on whether you use float16 or float32). This allows you to embed 50-100k documents per second on a cpu on a macbook. This reduction of course comes at a cost: distilled models are worse than their parent models. Even so, they are actually a lot better than large sets of conventional…
2024 · github.com
- 8AF
Hey HN — I’m Gaurav, one of the founders of Captions. We work on applied AI research for talking videos. Our foundation model, Lipdub, captures how humans speak, and matches full face movement to what’s being said. The model is zero-shot and can generate videos in under a minute, without person-specific training. Building on Lipdub, we’re releasing a few APIs that can generate and translate talking videos in bulk. Here are some ways they could be used: * Translating videos with matching lip movement * Creating personalized videos that include someone’s name or company, like what’s shown in…
2024 · captions.ai
- 9DR
The first ever AI peer reviewed research article just got approved. It’s kinda crazy how advanced AI have come to replace researchers. I've just been using Deep Research on ChatGPT and Perplexity a lot to write and research complex technical reports for my boss. He loves the reports and it has decreased my workload a ton but I still have some frustrations with it. None of them provide an API that gets me the same quality of output you would with the applications. I wanted something with more control on the LLMs, swappable with the reasoning new models that came out. Not just prompt →…
2025 · github.com
- 1001
Hey HN! I've been working on a side project to create an audio transcription API based on the OpenAI whisper model. Sign up link: https://whisperapi.com I tried to make the API really easy to use and get setup with. Also, because the Whisper model is so good, turns out I can offer the service for about 75% cheaper than what seems like the industry average. I'm always looking to make improvements, so would appreciate any feedback anyone has!
2022 · whisperapi.com
- 11AA
Hi guys, For a few months now I've been working on a web GUI to build, visualise, train and share deep neural models. It's currently reaching a state where opening it for Beta release make sense. Currently the tool support: - Fully connected and Convolutional architecture - Cloud and local, saving / loading of models - Edit / delete layers - Visualise Convolutional layers filters - Freeze / Unfreeze layers - More datasets: Fashion MNIST, QuickDraw(10 and 30) The editor can be found here: https://aifiddle.io. Your feedback, ideas, suggestions are greatly useful, so…
2019
- 12DA
hey HN, happy Monday! I got sick of receiving personal messages that were clearly written by AI, so I vibe-coded a pastebin-like service that allows authors to electively prove they wrote something themselves. how it works: - typing rhythm analysis: analyzing your keystroke patterns to measure how human-like they are, based on rhythm, speed, variance, etc. - audio verification: optionally, corroborating the audio from your microphone with your keystrokes as you type, to get an "audio verification" badge. in an increasingly AI-generated world, I think it's very important for us humans to be…
Feb 2026 · droptheslop.ai
- 13MC
Hey everyone! Many of you might have come across the Mamba paper a few days ago, which introduced an LLM based on a state space model architecture. The Mamba architecture is quite useful as its complexity scales subquadratically with input length and is therefore way more efficient than transformer models: https://github.com/state-spaces/mamba I got really excited about the paper, so I decided to fine-tune the model on a chat dataset. It turns that this actually worked quite well! The model is quite suitable for casual chatting, which honestly surprised me given that it…
2023 · github.com
- 14FT
I built a simple text-to-speech converter at texttospeech.site Free tier: 10 generations/day, standard voices, no account needed. Pro tier: Neural2 voices, 2000 chars, downloadable MP3s. Stack: Next.js, Google Cloud TTS API, Vercel. The $2 domain was an SEO experiment after my speechtotext.xyz satellite drove 22% of traffic to my main product. Curious if exact-match keyword domains still work for TTS searches. Feedback welcome — especially on voice quality and UX.
Jan 2026 · texttospeech.site
- 15IM
Hello all, I made a small transcription app for your Mac based on OpenAI’s Whisper. Would love some feedback. My plan is to make it easy to load weights from any fine-tuned whisper model to enable specialized dictation for any subfield. It’s still early in development. Thanks!
2023 · twitter.com
- 16MM
Hi HN! We (Thomas and Stéphan, hello!) recently released Model2Vec, a Python library for distilling any sentence transformer into a small set of static embeddings. This makes inference with such a model up to 500x faster, and reduces model size by a factor of 15 (7.5M params or 15/30MB on disk, depending on whether you use float16 or float32). This reduction of course comes at a cost: distilled models are a lot worse than their parent models. Even so, they are actually a lot better than large sets of conventional static embeddings, such as GLoVe or word2vec-based models, which are many…
2024 · github.com
- 17AA
An all-in-one blog for learning LLM ins and outs: tokenize, attention, PE, and more Project I've been diving deep into the internals of Large Language Models (LLMs) and started documenting my findings. My blog covers topics like: Tokenization techniques (e.g., BBPE) Attention mechanism (e.g. MHA, MQA, MLA) Positional encoding and extrapolation (e.g. RoPE, NTK-aware interpolation, YaRN) Architecture details of models like QWen, LLaMA Training methods including SFT and Reinforcement Learning If you're interested in the nuts and bolts of LLMs, feel free to check it out:…
2025 · comfyai.app
- 18IE
Hey HN, when building ML systems for industrial AI, we have learned that data inspection is critical during the ML development process. We are also big fans of the Hugging Face ecosystem. That is why we built an integration to our data exploration tool Spotlight that allows you to interactively explore Hugging Face datasets with one line of code. Spotlight lets you leverage model results such as predictions and embeddings to gain a deeper understanding in data segments and model failure modes. Currently, many many NLP, CV, Audio and multimodal datasets are supported both locally and on the…
2023 · huggingface.co
- 19TT
Hey HN, this is deepan from trulytyped (https://trulytyped.com). I am building a document writing app which makes it extremely easy to figure out how a document was created. Now that any text can be AI generated, how do you tell if something was actually generated or composed. It is impossible to detect AI after a piece of text has been generated. No amount of watermarking, linguistic checks or vibe checks work consistently. The AI detectors that schools and journals use are easy to bypass. Why do we need to solve this problem - First of all, this is not an anti-AI stance. I have…
May 2026 · trulytyped.com
- 20AA
We’re a group of electronic music artists and PhD researchers working in AI music. We've been working on a project to help us discover new music in a more objective way. It's called Speak Music: https://speakmusic.sonophase.com/ We’ve trained an AI model to understand the correspondence between music and language. The model combines a machine listening and audio signal processing with transformers for text embeddings. Once trained, we index a huge catalogue of unseen audio, ensuring that the search system can efficiently scale to millions of tracks. At the moment, our model is…
2023 · speakmusic.sonophase.com
- 21SC
Hey HN Community! We're excited to introduce Spine, a tool we've been developing for the past month that aims to streamline the process of building and sharing AI-driven natural language interfaces like ChatGPT for various data formats. With Spine, you can: * Upload numerous data types such as websites, PDFs, docs, PowerPoints, CSVs, audio files, YouTube videos, and more * Navigate lots of data and get in-depth referenced results with our hybrid search * Built-in Feedback: Craft accurate, personalized & adaptive experiences -- we will update the search models and eventually your LLM with the…
2023 · getspine.ai
- 22FS
We needed a speaker diarization solution that could run every few seconds alongside transcription on iOS and macOS. But native Swift support was either limited or locked behind paid licenses. Since diarization is a common need in speech-to-text workflows, we decided to open source our work and give back to the community. We initially tried sherpa-onnx, which works, but running both diarization and transcription models slowed down older devices. CPU-only inference just isn’t ideal for near real-time workloads, so we wanted the option to offload segmentation and speaker embedding to the GPU or…
2025 · github.com
- 23AV
I found myself jumping between ChatGPT, tabs, and docs, but never building real understanding. This is my attempt at fixing that — for researchers, curious readers, and lifelong learners. Would love your thoughts on the interface, and whether this would be useful in your own work or otherwise.
2025 · proread.ai
- 24OS
Hey HN, it’s Russ - cofounder of LiveKit. An open source stack for building realtime AI applications. We’re sharing our first homegrown AI model for turn detection. Here’s a live demo: https://cerebras.vercel.app/ Voice AI has come a long way in the last year. We now have end-to-end systems that can generate a response to user input in 300-500ms — human level speeds! As latency reduces, a common problem that surfaces is the LLM responds too quickly. Any time there’s a short pause in a user’s speech, it ends up interrupting them. This is largely due to how voice AI applications…
2024
Ranked by how close each launch is in meaning, then by votes. Refine with a description →