linto-ai / whisper-timestamped

Multilingual Automatic Speech Recognition with word-level timestamps and confidence
GNU Affero General Public License v3.0
2.06k stars 156 forks source link
asr attention-is-all-you-need attention-mechanism attention-model attention-network attention-seq2seq attention-visualization deep-learning machine-learning multilingual-models python python3 pytorch speaker-diarization speech speech-processing speech-recognition speech-to-text transformers whisper

whisper-timestamped

Multilingual Automatic Speech Recognition with word-level timestamps and confidence.

Description

Whisper is a set of multi-lingual, robust speech recognition models trained by OpenAI that achieve state-of-the-art results in many languages. Whisper models were trained to predict approximate timestamps on speech segments (most of the time with 1-second accuracy), but they cannot originally predict word timestamps. This repository proposes an implementation to predict word timestamps and provide a more accurate estimation of speech segments when transcribing with Whisper models. Besides, a confidence score is assigned to each word and each segment.

The approach is based on Dynamic Time Warping (DTW) applied to cross-attention weights, as demonstrated by this notebook by Jong Wook Kim. There are some additions to this notebook:

whisper-timestamped is an extension of the openai-whisper Python package and is meant to be compatible with any version of openai-whisper. It provides more efficient/accurate word timestamps, along with those additional features:

Notes on other approaches

An alternative relevant approach to recovering word-level timestamps involves using wav2vec models that predict characters, as successfully implemented in whisperX. However, these approaches have several drawbacks that are not present in approaches based on cross-attention weights such as whisper_timestamped. These drawbacks include:

An alternative approach that does not require an additional model is to look at the probabilities of timestamp tokens estimated by the Whisper model after each (sub)word token is predicted. This was implemented, for instance, in whisper.cpp and stable-ts. However, this approach lacks robustness because Whisper models have not been trained to output meaningful timestamps after each word. Whisper models tend to predict timestamps only after a certain number of words have been predicted (typically at the end of a sentence), and the probability distribution of timestamps outside this condition may be inaccurate. In practice, these methods can produce results that are totally out-of-sync on some periods of time (we observed this especially when there is jingle music). Also, the timestamp precision of Whisper models tends to be rounded to 1 second (as in many video subtitles), which is too inaccurate for words, and reaching better accuracy is tricky.

Installation

First installation

Requirements:

You can install whisper-timestamped either by using pip:

pip3 install whisper-timestamped

or by cloning this repository and running installation:

git clone https://github.com/linto-ai/whisper-timestamped
cd whisper-timestamped/
python3 setup.py install

Additional packages that might be needed

If you want to plot alignment between audio timestamps and words (as in this section), you also need matplotlib:

pip3 install matplotlib

If you want to use VAD option (Voice Activity Detection before running Whisper model), you also need torchaudio and onnxruntime:

pip3 install onnxruntime torchaudio

If you want to use finetuned Whisper models from the Hugging Face Hub, you also need transformers:

pip3 install transformers

Docker

A docker image of about 9GB can be built using:

git clone https://github.com/linto-ai/whisper-timestamped
cd whisper-timestamped/
docker build -t whisper_timestamped:latest .

Light installation for CPU

If you don't have a GPU (or don't want to use it), then you don't need to install the CUDA dependencies. You should then just install a light version of torch before installing whisper-timestamped, for instance as follows:

pip3 install \
     torch==1.13.1+cpu \
     torchaudio==0.13.1+cpu \
     -f https://download.pytorch.org/whl/torch_stable.html

A specific docker image of about 3.5GB can also be built using:

git clone https://github.com/linto-ai/whisper-timestamped
cd whisper-timestamped/
docker build -t whisper_timestamped_cpu:latest -f Dockerfile.cpu .

Upgrade to the latest version

When using pip, the library can be updated to the latest version using:

pip3 install --upgrade --no-deps --force-reinstall git+https://github.com/linto-ai/whisper-timestamped

A specific version of openai-whisper can be used by running, for example:

pip3 install openai-whisper==20230124

Usage

Python

In Python, you can use the function whisper_timestamped.transcribe(), which is similar to the function whisper.transcribe():

import whisper_timestamped
help(whisper_timestamped.transcribe)

The main difference with whisper.transcribe() is that the output will include a key "words" for all segments, with the word start and end position. Note that the word will include punctuation. See the example below.

Besides, the default decoding options are different to favour efficient decoding (greedy decoding instead of beam search, and no temperature sampling fallback). To have same default as in whisper, use beam_size=5, best_of=5, temperature=(0.0, 0.2, 0.4, 0.6, 0.8, 1.0).

There are also additional options related to word alignement.

In general, if you import whisper_timestamped instead of whisper in your Python script and use transcribe(model, ...) instead of model.transcribe(...), it should do the job:

import whisper_timestamped as whisper

audio = whisper.load_audio("AUDIO.wav")

model = whisper.load_model("tiny", device="cpu")

result = whisper.transcribe(model, audio, language="fr")

import json
print(json.dumps(result, indent = 2, ensure_ascii = False))

Note that you can use a finetuned Whisper model from HuggingFace or a local folder by using the load_model method of whisper_timestamped. For instance, if you want to use whisper-large-v2-nob, you can simply do the following:

import whisper_timestamped as whisper

model = whisper.load_model("NbAiLab/whisper-large-v2-nob", device="cpu")

# ...

Command line

You can also use whisper_timestamped on the command line, similarly to whisper. See help with:

whisper_timestamped --help

The main differences with whisper CLI are:

An example command to process several files using the tiny model and output the results in the current folder, as would be done by default with whisper, is as follows:

whisper_timestamped audio1.flac audio2.mp3 audio3.wav --model tiny --output_dir .

Note that you can use a fine-tuned Whisper model from HuggingFace or a local folder. For instance, if you want to use the whisper-large-v2-nob model, you can simply do the following:

whisper_timestamped --model NbAiLab/whisper-large-v2-nob <...>

Utility Functions

In addition to the main transcribe function, whisper-timestamped provides some utility functions:

remove_non_speech

Remove non-speech segments from audio using Voice Activity Detection (VAD).

from whisper_timestamped import remove_non_speech

audio_speech, segments, convert_timestamps = remove_non_speech(audio, vad="silero")

load_model

Load a Whisper model from a given name or path, including support for fine-tuned models from HuggingFace.

from whisper_timestamped import load_model

model = load_model("NbAiLab/whisper-large-v2-nob", device="cpu")

Plot of word alignment

Note that you can use the plot_word_alignment option of the whisper_timestamped.transcribe() Python function or the --plot option of the whisper_timestamped CLI to see the word alignment for each segment.

Example alignement

Example output

The output of whisper_timestamped.transcribe() function is a python dictionary, which can be viewed in JSON format using the CLI.

The JSON schema can be seen in tests/json_schema.json.

Here is an example output:

whisper_timestamped AUDIO_FILE.wav --model tiny --language fr
{
  "text": " Bonjour! Est-ce que vous allez bien?",
  "segments": [
    {
      "id": 0,
      "seek": 0,
      "start": 0.5,
      "end": 1.2,
      "text": " Bonjour!",
      "tokens": [ 25431, 2298 ],
      "temperature": 0.0,
      "avg_logprob": -0.6674491882324218,
      "compression_ratio": 0.8181818181818182,
      "no_speech_prob": 0.10241222381591797,
      "confidence": 0.51,
      "words": [
        {
          "text": "Bonjour!",
          "start": 0.5,
          "end": 1.2,
          "confidence": 0.51
        }
      ]
    },
    {
      "id": 1,
      "seek": 200,
      "start": 2.02,
      "end": 4.48,
      "text": " Est-ce que vous allez bien?",
      "tokens": [ 50364, 4410, 12, 384, 631, 2630, 18146, 3610, 2506, 50464 ],
      "temperature": 0.0,
      "avg_logprob": -0.43492694334550336,
      "compression_ratio": 0.7714285714285715,
      "no_speech_prob": 0.06502953916788101,
      "confidence": 0.595,
      "words": [
        {
          "text": "Est-ce",
          "start": 2.02,
          "end": 3.78,
          "confidence": 0.441
        },
        {
          "text": "que",
          "start": 3.78,
          "end": 3.84,
          "confidence": 0.948
        },
        {
          "text": "vous",
          "start": 3.84,
          "end": 4.0,
          "confidence": 0.935
        },
        {
          "text": "allez",
          "start": 4.0,
          "end": 4.14,
          "confidence": 0.347
        },
        {
          "text": "bien?",
          "start": 4.14,
          "end": 4.48,
          "confidence": 0.998
        }
      ]
    }
  ],
  "language": "fr"
}

If the language is not specified (e.g. without option --language fr in the CLI) you will find an additional key with the language probabilities:

{
  ...
  "language": "fr",
  "language_probs": {
    "en": 0.027954353019595146,
    "zh": 0.02743500843644142,
    ...
    "fr": 0.9196318984031677,
    ...
    "su": 3.0119704064190955e-08,
    "yue": 2.2565967810805887e-05
  }
}

API Reference

Main Transcription Function

transcribe_timestamped(model, audio, **kwargs)

Transcribe audio using a Whisper model and compute word-level timestamps.

Parameters:
Returns:

A dictionary containing:

Exceptions:
Notes:

Utility Functions

remove_non_speech(audio, **kwargs)

Remove non-speech segments from audio using Voice Activity Detection (VAD).

Parameters:
Returns:

A tuple containing:

  1. torch.Tensor: Audio with non-speech segments removed
  2. List[Tuple[float, float]]: List of (start, end) timestamps for speech segments
  3. Callable: Function to convert timestamps from the new audio to the original audio
Exceptions:
Notes:

load_model(name, device=None, backend="openai-whisper", download_root=None, in_memory=False)

Load a Whisper model from a given name or path.

Parameters:
Returns:

The loaded Whisper model.

Exceptions:
Notes:

get_alignment_heads(model, max_top_layer=3)

Get the alignment heads for the given model.

Parameters:
Returns:

A sparse tensor representing the alignment heads.

Notes:

File Writing Functions

The following functions are available for writing transcripts to various file formats:

write_csv(transcript, file, sep=",", text_first=True, format_timestamps=None, header=False)

Write transcript data to a CSV file.

Parameters:
Exceptions:
Notes:

write_srt(transcript, file)

Write transcript data to an SRT (SubRip Subtitle) file.

Parameters:
Exceptions:
Notes:

write_vtt(transcript, file)

Write transcript data to a VTT (WebVTT) file.

Parameters:
Exceptions:
Notes:

write_tsv(transcript, file)

Write transcript data to a TSV (Tab-Separated Values) file.

Parameters:
Exceptions:
Notes:

Options that may improve results

Here are some options that are not enabled by default but might improve results.

Accurate Whisper transcription

As mentioned earlier, some decoding options are disabled by default to offer better efficiency. However, this can impact the quality of the transcription. To run with the options that have the best chance of providing a good transcription, use the following options.

Running Voice Activity Detection (VAD) before sending to Whisper

Whisper models can "hallucinate" text when given a segment without speech. This can be avoided by running VAD and gluing speech segments together before transcribing with the Whisper model. This is possible with whisper-timestamped.

By default, the VAD method used is silero. But other methods are available, such as earlier versions of silero, or auditok. Those methods were introduced because latest versions of silero VAD can have a lot of false alarms on some audios (speech detected on silence).

In order to watch the VAD results, you can use the --plot option of the whisper_timestamped CLI, or the plot_word_alignment option of the whisper_timestamped.transcribe() Python function. It will show the VAD results on the input audio signal as following (x-axis is time in seconds): vad="silero:v4.0" vad="silero:v3.1" vad="auditok"
Example VAD Example VAD Example VAD

Detecting disfluencies

Whisper models tend to remove speech disfluencies (filler words, hesitations, repetitions, etc.). Without precautions, the disfluencies that are not transcribed will affect the timestamp of the following word: the timestamp of the beginning of the word will actually be the timestamp of the beginning of the disfluencies. whisper-timestamped can have some heuristics to avoid this.

Acknowledgment

Citations

If you use this in your research, please cite the repo:

@misc{lintoai2023whispertimestamped,
  title={whisper-timestamped},
  author={Louradour, J{\'e}r{\^o}me},
  journal={GitHub repository},
  year={2023},
  publisher={GitHub},
  howpublished = {\url{https://github.com/linto-ai/whisper-timestamped}}
}

as well as the OpenAI Whisper paper:

@article{radford2022robust,
  title={Robust speech recognition via large-scale weak supervision},
  author={Radford, Alec and Kim, Jong Wook and Xu, Tao and Brockman, Greg and McLeavey, Christine and Sutskever, Ilya},
  journal={arXiv preprint arXiv:2212.04356},
  year={2022}
}

and this paper for Dynamic-Time-Warping:

@article{JSSv031i07,
  title={Computing and Visualizing Dynamic Time Warping Alignments in R: The dtw Package},
  author={Giorgino, Toni},
  journal={Journal of Statistical Software},
  year={2009},
  volume={31},
  number={7},
  doi={10.18637/jss.v031.i07}
}