Add processors for downloading HiFiTTS2 dataset

[rlangman]

Adds processors needed to download HiFiTTS-2. The input to the processors will be two files (e.g. manifest_22khz and chapters_22khz) that the user downloads from another location:

Example command:

python /home/NeMo-speech-data-processor/main.py \
    --config-path="/home/NeMo-speech-data-processor/dataset_configs/english/hifitts2" \
    --config-name="config_22khz.yaml" \
    workspace_dir="/home/hifitts2" \
    max_workers=8

This PR also contains a generic processor to estimate bandwidth of audio, which was used in creating HiFiTTS-2. It is not part of the downloading pipeline itself, as it is already precomputed and provided in the dataset manifest.

Example command:

python /home/NeMo-speech-data-processor/main.py \
    --config-path="/home/NeMo-speech-data-processor/dataset_configs/english/hifitts2" \
    --config-name="config_bandwidth.yaml" \
    workspace_dir="/home/hifitts2" \
    audio_dir_name="audio_22khz" \
    input_manifest_file="manifest_22khz.json" \
    max_workers=8

[karpnv]

Grea work! Please add 2 end-to-end tests for 2 new pipelines and mention new processors in
https://github.com/NVIDIA/NeMo-speech-data-processor/blob/main/docs/src/sdp/api.rst

[karpnv]

Since we usually skip a download processor in the end-to-end test, you can just test RemovedFailedChapters processor in the unit test

[rlangman]

Grea work! Please add 2 end-to-end tests for 2 new pipelines and mention new processors in https://github.com/NVIDIA/NeMo-speech-data-processor/blob/main/docs/src/sdp/api.rst

I tried to add an end to end test, and more documentation. Let me know if there are any issues, or if I need to upload test data.

I also added a generic bandwidth estimation script, which was used when creating the dataset (but running it during the dataset download is not needed).

[Jorjeous]

"/home/runner/work/NeMo-speech-data-processor/NeMo-speech-data-processor/test_data/english/hifitts2/manifest_filtered_22khz.json"
This file is being used but seems that tests starts before it's being created, lets fix this

[Jorjeous]

Ways to solve:
1)change test config so the file is being created firstly
2) you can read this file from code itself
3) or upload to aws and download in process

[Jorjeous]

Docs ok, errors expected as pages not exist yet

[Jorjeous]

Also lacking files:
/home/runner/work/NeMo-speech-data-processor/NeMo-speech-data-processor/test_data/english/hifitts2/test_data_reference_bandwidth.json
/home/runner/work/NeMo-speech-data-processor/NeMo-speech-data-processor/test_data/english/hifitts2/manifest_filtered_22khz.json
/home/runner/work/NeMo-speech-data-processor/NeMo-speech-data-processor/test_data/english/hifitts2/errors_22khz.json'
/home/runner/work/NeMo-speech-data-processor/NeMo-speech-data-processor/test_data/english/hifitts2/manifest_22khz.json

[Jorjeous]

Could you please rewrite class descriptions in this format? (codeblock with example at the end)

class CreateInitialManifestYTC(BaseParallelProcessor):
"""A processor class for creating initial manifest files for a TTS dataset.

It takes a manifest file containing audio file paths and resamples them to a target
sample rate and format, while creating a new manifest file with the updated paths.

Args:
    input_format (str): Format of the input audio files
    resampled_audio_dir (str): Directory where resampled audio files will be saved
    target_sample_rate (int): Desired sample rate for the output audio files
    target_format (str): Desired format for the output audio files
    target_nchannels (int): Desired number of channels for the output audio files

Returns:
    The same data as in the input manifest, but with resampled audio files and
    updated audio file paths.

Example:
    .. code-block:: yaml

        - _target_: sdp.processors.datasets.ytc.create_initial_manifest.CreateInitialManifestYTC
          input_manifest_file: ${workspace_dir}/manifest.json
          output_manifest_file: ${workspace_dir}/manifest_resampled.json
"""

[Jorjeous]

When mentioned above comments is addressed every thing is good to go!
Nice!

[rlangman]

Could you please rewrite class descriptions in this format? (codeblock with example at the end)

class CreateInitialManifestYTC(BaseParallelProcessor): """A processor class for creating initial manifest files for a TTS dataset.

It takes a manifest file containing audio file paths and resamples them to a target
sample rate and format, while creating a new manifest file with the updated paths.

Args:
    input_format (str): Format of the input audio files
    resampled_audio_dir (str): Directory where resampled audio files will be saved
    target_sample_rate (int): Desired sample rate for the output audio files
    target_format (str): Desired format for the output audio files
    target_nchannels (int): Desired number of channels for the output audio files

Returns:
    The same data as in the input manifest, but with resampled audio files and
    updated audio file paths.

Example:
    .. code-block:: yaml

        - _target_: sdp.processors.datasets.ytc.create_initial_manifest.CreateInitialManifestYTC
          input_manifest_file: ${workspace_dir}/manifest.json
          output_manifest_file: ${workspace_dir}/manifest_resampled.json
"""

Added Example section in documentation.

[rlangman]

Also lacking files: /home/runner/work/NeMo-speech-data-processor/NeMo-speech-data-processor/test_data/english/hifitts2/test_data_reference_bandwidth.json /home/runner/work/NeMo-speech-data-processor/NeMo-speech-data-processor/test_data/english/hifitts2/manifest_filtered_22khz.json /home/runner/work/NeMo-speech-data-processor/NeMo-speech-data-processor/test_data/english/hifitts2/errors_22khz.json' /home/runner/work/NeMo-speech-data-processor/NeMo-speech-data-processor/test_data/english/hifitts2/manifest_22khz.json

I have setup the data on S3 so it passes when I run locally. Will see if it passes in the automated PR tests.

[Jorjeous]

LGTM