Skip to content

Commit

Permalink
Add note on command-line flags.
Browse files Browse the repository at this point in the history
PiperOrigin-RevId: 706691578
  • Loading branch information
agutkin committed Dec 16, 2024
1 parent 14dd9a1 commit b58cebd
Show file tree
Hide file tree
Showing 2 changed files with 10 additions and 2 deletions.
2 changes: 2 additions & 0 deletions protoscribe/evolution/make_html_main.py
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,8 @@

r"""Constructs an HTML page to view the proposed spellings with glyph images.
See `make_html.py` for the definitions of relevant flags.
Example:
--------
EXPERIMENT_DIR=...
Expand Down
10 changes: 8 additions & 2 deletions protoscribe/speech/audio_tokenizer.py
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,8 @@

"""Audio tokenization interfaces."""

import logging

from protoscribe.speech import abstract_audio_tokenizer

AudioTokenizer = abstract_audio_tokenizer.AudioTokenizer
Expand All @@ -31,12 +33,16 @@ def get_tokenizer(
model_config_name_or_path: Name of the model configuration or path of
the model. This is implementation-specific.
sample_rate: Sampling rate in Hz.
has_quantizer: True if the model has quantizer. In this case it should be
possible to retrieve discrete tokens in addition to the embeddings.
has_quantizer: True if the model has quantizer. In this case it should be
possible to retrieve discrete tokens in addition to the embeddings.
normalize_embeddings: Use embeddings as is (default) or normalize them.
Returns:
Audio tokenizer instance.
"""
logging.info(
"Model: '%s', sample_rate: %d, normalize embeddings: %d.",
model_config_name_or_path, sample_rate, normalize_embeddings
)
# No audio tokenizers have been implemented yet.
return None

0 comments on commit b58cebd

Please sign in to comment.