Deepgram VS Kaldi ASR

Deepgram Reviews

1. suz11    

   10 months ago
   Fast

   It's fast - but for an API, not the fastest speech-to-text. For a long while I hadn't done research and trusted them. Then tried Whisper and Picovoice. On-device latency is nothing comparable with cloud APIs. If latency is important go with Whisper or Picovoice. If customization is also important go with Picovoice.

   don't get me wrong it's still faster than amazon, Microsoft or Assemblyai

   🏁 Competitors: Picovoice.ai, Google Cloud Speech API, AssemblyAI, Speechmatics, Microsoft Azure, Amazon

   👍 Pros:    Fast|Fast sales|A free demo|Supports multiple languages

   👎 Cons:    Slow support for free customers|Limited free version

Kaldi ASR Reviews

Deepgram mentions (27)

• Building a Live Proctoring System to Detect Multiple Speakers

  Lastly, we will be using Deepgram Audio Diarization APIs to get speaker details from a sample audio clip. - Source: dev.to / 4 months ago

• How do I begin building AI tools for myself?

  There are other AI-powered APIs out there to consider, too. For example, Deepgram can be used to transcribe audio (better than Whisper, offered by OpenAI), ElevenLabs can be used to generate speech from text (including using custom voices, which OpenAI's TTS can't currently do), etc. Depending on what you're trying to make, a combination of these services may be what you need. In any case, Python is going to be... Source: 5 months ago

• Use Langchain, Deepgram, and Mistral 7B to Build a Youtube Video Summarization App

  This guide delves deep into the world of YouTube video summarization, harnessing the power of cutting-edge technologies including Deepgram for superior audio transcription, Langchain for harvesting the power of the LLM, and Mistral 7B, a state-of-the-art and open-source LLM. - Source: dev.to / 6 months ago

• Create a realtime closed captioning solution with Deepgram and Ably

  Historically it's been challenging to provide closed captioning for live experiences, be it a live interview, a sports game with commentary, or a livestream. But Deepgram's AI tooling has changed this, allowing users to easily convert realtime streams of audio into accurate transcripts. - Source: dev.to / 6 months ago

• Introducing Daily's AI-powered Clinical Notes API

  To deliver the best possible output, we’ve tested a wide range of AI models and supporting tools. Today, we’re producing industry-leading clinical notes text by combining Deepgram’s highly capable transcription models, ScienceIO’s healthcare-specific AI for processing medical data, and the Microsoft Azure OpenAI GPT-4 large language model. We’ve worked closely with all three teams to bring this new capability to... - Source: dev.to / 8 months ago

Kaldi ASR mentions (3)

• C++ for machine learning

  Additionally, C++ may be used for extremely high levels of optimization even for cloud-based ML. Dlib and Kaldi are C++ libraries used as dependencies in Python codebases for computer vision and audio processing, for example. So if your application requires you to customize any functions similar to those libraries, then you'll need C++ knowhow. Source: over 1 year ago

• SOTA speech to text framework?

  I'm not how sure it stacks with recent state of art, but Kaldi toolkit (https://github.com/kaldi-asr/kaldi) used to be popular for building all kinds of practical integrations and experiments for speech recognition. Source: about 2 years ago

• Nerd-dictation, hackable speech to text on Linux

  Vosk-api isn't an SST engine itself, it is built using the Kaldi speech recognition toolkit (https://github.com/kaldi-asr/kaldi) and nicely implements and packages an API for Kaldi chain/LF-MMI models. - Source: Hacker News / over 2 years ago