Difference between revisions of "Speech Recognition"
m |
m |
||
| Line 16: | Line 16: | ||
* [[Attention]] Mechanism ...[[Transformer]] Model ...[[Generative Pre-trained Transformer (GPT)]] | * [[Attention]] Mechanism ...[[Transformer]] Model ...[[Generative Pre-trained Transformer (GPT)]] | ||
* [[Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM)]] | * [[Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM)]] | ||
| + | * [[Generative AI]] ... [[OpenAI]]'s [[ChatGPT]] ... [[Perplexity]] ... [[Microsoft]]'s [[BingAI]] ... [[You]] ...[[Google]]'s [[Bard]] | ||
* [http://www.theverge.com/2022/9/23/23367296/openai-whisper-transcription-speech-recognition-open-source Iused] | * [http://www.theverge.com/2022/9/23/23367296/openai-whisper-transcription-speech-recognition-open-source Iused] | ||
Revision as of 15:34, 8 March 2023
YouTube search... ...Google search
- Capabilities
- End-to-End Speech
- Text to Speech
- Synthesize Speech
- Assistants ... Hybrid Assistants ... Agents ... Negotiation
- Natural Language Processing (NLP) ...Generation ...LLM ...Tools & Services
- Attention Mechanism ...Transformer Model ...Generative Pre-trained Transformer (GPT)
- Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM)
- Generative AI ... OpenAI's ChatGPT ... Perplexity ... Microsoft's BingAI ... You ...Google's Bard
- Iused
Whisper Automatic Speech Recognition Service (ASR)
YouTube search... ...Google search
Whisper is an Automatic Speech Recognition Service (ASR) by OpenAI trained on 680,000 hours of multilingual and multitask supervised data collected from the web. 'We’ve trained and are open-sourcing a neural net called Whisper that approaches human level robustness and accuracy on English speech recognition.' - OpenAI
