useSpeechToText

With the latest v0.3.0 release we introduce a new hook - useSpeechToText. Speech to text is a task that allows to transform spoken language to written text. It is commonly used to implement features such as transcription or voice assistants. As of now, all supported STT models run on the XNNPACK backend.

info

Currently, we do not support direct microphone input streaming to the model. Instead, in v0.3.0, we provide a way to transcribe an audio file.

caution

It is recommended to use models provided by us, which are available at our Hugging Face repository. You can also use constants shipped with our library

Reference

import { useSpeechToText } from 'react-native-executorch';

const { transcribe, error, loadAudio } = useSpeechToText({
  modelName: 'moonshine',
});

const audioUrl = ...; // URL with audio to transcribe

await loadAudio(audioUrl);
const transcription = await transcribe();
if (error) {
  console.log(error);
} else {
  console.log(transcription);
}

Streaming

Given that STT models can process audio no longer than 30 seconds, there is a need to chunk the input audio. Chunking audio may result in cutting speech mid-sentence, which might be hard to understand for the model. To make it work, we employed an algorithm (adapted for mobile devices from whisper-streaming) that uses overlapping audio chunks. This might introduce some overhead, but allows for processing audio inputs of arbitrary length.

Arguments

modelName A literal of "moonshine" | "whisper" which serves as an identifier for which model should be used.

encoderSource? A string that specifies the location of a .pte file for the encoder. For further information on passing model sources, check out Loading Models. Defaults to constants for given model.

decoderSource? Analogous to the encoderSource, this takes in a string which is a source for the decoder part of the model. Defaults to constants for given model.

tokenizerSource? A string that specifies the location to the tokenizer for the model. This works just as the encoder and decoder do. Defaults to constants for given model.

overlapSeconds? Specifies the length of overlap between consecutive audio chunks (expressed in seconds).

windowSize? Specifies the size of each audio chunk (expressed in seconds).

Returns

Field	Type	Description
transcribe	(input?: number[]) => Promise<string>	Starts a transcription process for a given input array, which should be a waveform at 16kHz. When no input is provided, it uses an internal state which is set by calling loadAudio. Resolves a promise with the output transcription when the model is finished.
loadAudio	(url: string) => void	Loads audio file from given url. It sets an internal state which serves as an input to transcribe().
error	Error | undefined	Contains the error message if the model failed to load.
sequence	string	This property is updated with each generated token. If you're looking to obtain tokens as they're generated, you should use this property.
isGenerating	boolean	Indicates whether the model is currently processing an inference.
isReady	boolean	Indicates whether the model has successfully loaded and is ready for inference.
downloadProgress	number	Tracks the progress of the model download process.

Running the model

Before running the model's transcribe method be sure to obtain waveform of the audio You wish to transcribe. You can either use loadAudio method to load audio from a url and save it in model's internal state or obtain the waveform on your own (remember to use sampling rate of 16kHz!). In the latter case just pass the obtained waveform as argument to the transcribe method which returns a promise resolving to the generated tokens when successful. If the model fails during inference the error property contains details of the error. If you want to obtain tokens in a streaming fashion, you can also use the sequence property, which is updated with each generated token, similar to the useLLM hook.

Example

import { Button, Text } from 'react-native';
import { useSpeechToText } from 'react-native-executorch';

function App() {
  const { loadAudio, transcribe, sequence, error } = useSpeechToText({
    modelName: 'whisper',
  });

  const audioUrl = ...; // URL with audio to transcribe

  return (
    <View>
      <Button
        onPress={async () => {
          await loadAudio(audioUrl);
          await transcribe();
        }
        title="Transcribe"
      />
      <Text>{error ? error : sequence}</Text>
    </View>
  );
}

Supported models

Model	Language
Whisper tiny.en	English
Moonshine tiny	English

Benchmarks

Model size

Model	XNNPACK [MB]
WHISPER_TINY	231.0
MOONSHINE_TINY	148.9

Memory usage

Model	Android (XNNPACK) [MB]	iOS (XNNPACK) [MB]
WHISPER_TINY	900	600
MOONSHINE_TINY	650	560