WhisperX<T>
Models & Types · AiDotNet.SpeechRecognition.WhisperFamily
WhisperX (Bain et al. 2023): Whisper transcription with voice-activity cut-and-merge and word-level timing.
Inheritance: NeuralNetworkBase<T> → AudioNeuralNetworkBase<T> → WhisperX<T>
Implements: ISpeechRecognizer<T>
For Beginners
WhisperX listens to 16 kHz audio in 30-second windows turned into a log-mel spectrogram, encodes each window with a Transformer, and a second Transformer writes the transcript token by token while reading the encoded audio.
How It Works
WhisperX transcribes with Whisper (the network here is large-v3) and adds word timing. The paper times words by forced alignment with a phoneme model; transcription here times them by dynamic time warping over Whisper’s own cross-attention, openai-whisper’s word_timestamps. Every forward reads the 30-second window openai-whisper builds (pad_or_trim); Predict returns the
decoder’s logits over the start sequence (start of transcript, language, transcribe, no timestamps), whose last row is
the first token’s distribution, and Train minimizes the teacher-forced cross-entropy of a transcript given as
token ids or as one distribution per token.
Example
var architecture = new NeuralNetworkArchitecture<float>(
inputType: InputType.OneDimensional,
taskType: NeuralNetworkTaskType.SpeechRecognition,
inputSize: 480000, outputSize: 51866);
var model = new WhisperX<float>(architecture);
var text = model.Transcribe(waveform).Text;
Constructors
| Constructor | Summary |
|---|---|
WhisperX(NeuralNetworkArchitecture<T> architecture, WhisperXOptions options, IGradientBasedOptimizer<T, Tensor<T>, Tensor<T>> optimizer) |
|
WhisperX(NeuralNetworkArchitecture<T> architecture, string modelPath, WhisperXOptions options) |
Properties
| Property | Summary |
|---|---|
IReadOnlyList<string> SupportedLanguages { get; } |
|
bool SupportsStreaming { get; } |
|
bool SupportsWordTimestamps { get; } |
Methods
| Method | Summary |
|---|---|
IReadOnlyDictionary<string, T> DetectLanguageProbabilities(Tensor<T> audio) |
|
IStreamingTranscriptionSession<T> StartStreamingSession(string language) |
|
ModelMetadata<T> GetModelMetadata() |
|
Task<TranscriptionResult<T>> TranscribeAsync(Tensor<T> audio, string language, bool? includeTimestamps, CancellationToken cancellationToken) |
|
Tensor<T> ForwardForTraining(Tensor<T> input) |
|
TranscriptionResult<T> Transcribe(Tensor<T> audio, string language, bool? includeTimestamps) |
Transcribes 16 kHz audio with openai-whisper’s transcribe: 30-second windows, temperature fallback, and word timing when timestamps are requested. |
string DetectLanguage(Tensor<T> audio) |