homebrewltd
/

Ichigo-llama3.1-s-instruct-v0.3-phase-2

Audio-Text-to-Text

sound language model

Model card Files Files and versions Community

jan-hq commited on Oct 4

Commit

53cad53

•

1 Parent(s): 482c8d0

Update README.md

Files changed (1) hide show

README.md +1 -0

README.md CHANGED Viewed

@@ -14,6 +14,7 @@ We have developed and released the family [Ichigo-llama3s](https://huggingface.c
 We expand the Semantic tokens experiment with WhisperVQ as a tokenizer for audio files from [homebrewltd/Ichigo-llama3.1-s-base-v0.3](https://huggingface.co/homebrewltd/Ichigo-llama3.1-s-base-v0.3) with nearly 1B tokens from [Instruction Speech WhisperVQ v3](homebrewltd/mixed-instruction-speech-whispervq-v3-full) dataset.
 This is the model checkpoint from step 7000. Due to some noise in the training data, it has an artificially higher score on the Speech Instruction benchmark.
 **Model developers** Homebrew Research.
 **Input** Text and sound.

 We expand the Semantic tokens experiment with WhisperVQ as a tokenizer for audio files from [homebrewltd/Ichigo-llama3.1-s-base-v0.3](https://huggingface.co/homebrewltd/Ichigo-llama3.1-s-base-v0.3) with nearly 1B tokens from [Instruction Speech WhisperVQ v3](homebrewltd/mixed-instruction-speech-whispervq-v3-full) dataset.
 This is the model checkpoint from step 7000. Due to some noise in the training data, it has an artificially higher score on the Speech Instruction benchmark.
 **Model developers** Homebrew Research.
 **Input** Text and sound.