@ondewo/s2t-client-typescript 4.0.0 → 5.3.0

package/README.md

@@ -1,20 +1,34 @@
-<p align="center">
-  <a href="https://www.ondewo.com">
-    <img alt="ONDEWO Logo" src="https://raw.githubusercontent.com/ondewo/ondewo-logos/master/github/ondewo_logo_github_2.png"/>
-  </a>
+<div align="center">
+  <table>
+    <tr>
+      <td>
+        <a href="https://ondewo.com/en/products/natural-language-understanding/">
+            <img width="400px" src="https://raw.githubusercontent.com/ondewo/ondewo-logos/master/ondewo_we_automate_your_phone_calls.png"/>
+        </a>
+      </td>
+    </tr>
+    <tr>
+       <td align="center">
+          <a href="https://www.linkedin.com/company/ondewo "><img width="40px" src="https://cdn-icons-png.flaticon.com/512/3536/3536505.png"></a>
+          <a href="https://www.facebook.com/ondewo"><img width="40px" src="https://cdn-icons-png.flaticon.com/512/733/733547.png"></a>
+          <a href="https://twitter.com/ondewo"><img width="40px" src="https://cdn-icons-png.flaticon.com/512/733/733579.png"> </a>
+          <a href="https://www.instagram.com/ondewo.ai/"><img width="40px" src="https://cdn-icons-png.flaticon.com/512/174/174855.png"></a>
+          <a href="https://badge.fury.io/js/%40ondewo%2Fs2t-client-typescript"><img src="https://badge.fury.io/js/%40ondewo%2Fs2t-client-typescript.svg" alt="npm version" height="32"></a>
+       </td>
+    </tr>
+  </table>
   <h1 align="center">
     ONDEWO S2T Client Typescript
   </h1>
-  <p align="center">
-    <a href="https://badge.fury.io/js/%40ondewo%2Fs2t-client-typescript"><img src="https://badge.fury.io/js/%40ondewo%2Fs2t-client-typescript.svg" alt="npm version" height="18"></a>
-  </p>
-</p>
+</div>
 
-# @ondewo/s2t-client-typescript
+## Overview
 
-`@ondewo/s2t-client-typescript` is a compiled version of the [ONDEWO S2T API](https://github.com/ondewo/ondewo-s2t-api). Here you can find the S2T API [documentation](https://ondewo.github.io/ondewo-s2t-api/).
+`@ondewo/s2t-client-typescript` is a compiled version of the [ONDEWO S2T API](https://github.com/ondewo/ondewo-s2t-api) using the [ONDEWO PROTO COMPILER](https://github.com/ondewo/ondewo-proto-compiler). Here you can find the S2T API [documentation](https://ondewo.github.io).
 
-## Installation
+ONDEWO APIs use [Protocol Buffers](https://github.com/google/protobuf) version 3 (proto3) as their Interface Definition Language (IDL) to define the API interface and the structure of the payload messages. The same interface definition is used for gRPC versions of the API in all languages.
+
+## Setup
 
 Using NPM:
 
@@ -22,6 +36,33 @@ Using NPM:
 npm i --save @ondewo/s2t-client-typescript
 ```
 
-## Documentation
+Using GitHub:
+
+```shell
+git clone https://github.com/ondewo/ondewo-s2t-client-typescript.git ## Clone repository
+cd ondewo-s2t-client-typescript                                      ## Change into repo-directoy
+make setup_developer_environment_locally                             ## Install dependencies
+```
+
+## Package structure
+
+```
+npm
+├── api
+│   ├── google
+│   │   └── protobuf
+│   │       ├── empty_pb.d.ts
+│   │       └── empty_pb.js
+│   └── ondewo
+│       └── s2t
+│           ├── speech-to-text_grpc_web_pb.d.ts
+│           ├── speech-to-text_grpc_web_pb.js
+│           ├── speech-to-text_pb.d.ts
+│           └── speech-to-text_pb.js
+├── LICENSE
+├── package.json
+├── public-api.d.ts
+├── public-api.js
+└── README.md
+```
 
-Read the full documentation on [GitHub](https://github.com/ondewo/ondewo-s2t-client-typescript).

package/api/ondewo/s2t/speech-to-text_pb.d.ts

@@ -7,8 +7,8 @@ export class TranscribeRequestConfig extends jspb.Message {
   getS2tPipelineId(): string;
   setS2tPipelineId(value: string): TranscribeRequestConfig;
 
-  getCtcDecoding(): CTCDecoding;
-  setCtcDecoding(value: CTCDecoding): TranscribeRequestConfig;
+  getDecoding(): Decoding;
+  setDecoding(value: Decoding): TranscribeRequestConfig;
 
   getLanguageModelName(): string;
   setLanguageModelName(value: string): TranscribeRequestConfig;
@@ -59,7 +59,7 @@ export class TranscribeRequestConfig extends jspb.Message {
 export namespace TranscribeRequestConfig {
   export type AsObject = {
     s2tPipelineId: string,
-    ctcDecoding: CTCDecoding,
+    decoding: Decoding,
     languageModelName: string,
     postProcessing?: PostProcessingOptions.AsObject,
     utteranceDetection?: UtteranceDetectionOptions.AsObject,
@@ -102,11 +102,20 @@ export class TranscriptionReturnOptions extends jspb.Message {
   getReturnAudio(): boolean;
   setReturnAudio(value: boolean): TranscriptionReturnOptions;
 
+  getReturnConfidenceScore(): boolean;
+  setReturnConfidenceScore(value: boolean): TranscriptionReturnOptions;
+
   getReturnAlternativeTranscriptions(): boolean;
   setReturnAlternativeTranscriptions(value: boolean): TranscriptionReturnOptions;
 
-  getReturnConfidenceScore(): boolean;
-  setReturnConfidenceScore(value: boolean): TranscriptionReturnOptions;
+  getReturnAlternativeTranscriptionsNr(): number;
+  setReturnAlternativeTranscriptionsNr(value: number): TranscriptionReturnOptions;
+
+  getReturnAlternativeWords(): boolean;
+  setReturnAlternativeWords(value: boolean): TranscriptionReturnOptions;
+
+  getReturnAlternativeWordsNr(): number;
+  setReturnAlternativeWordsNr(value: number): TranscriptionReturnOptions;
 
   getReturnWordTiming(): boolean;
   setReturnWordTiming(value: boolean): TranscriptionReturnOptions;
@@ -123,8 +132,11 @@ export namespace TranscriptionReturnOptions {
   export type AsObject = {
     returnStartOfSpeech: boolean,
     returnAudio: boolean,
-    returnAlternativeTranscriptions: boolean,
     returnConfidenceScore: boolean,
+    returnAlternativeTranscriptions: boolean,
+    returnAlternativeTranscriptionsNr: number,
+    returnAlternativeWords: boolean,
+    returnAlternativeWordsNr: number,
     returnWordTiming: boolean,
   }
 }
@@ -668,16 +680,19 @@ export namespace S2TDescription {
 }
 
 export class S2TInference extends jspb.Message {
-  getCtcAcousticModels(): CtcAcousticModels | undefined;
-  setCtcAcousticModels(value?: CtcAcousticModels): S2TInference;
-  hasCtcAcousticModels(): boolean;
-  clearCtcAcousticModels(): S2TInference;
+  getAcousticModels(): AcousticModels | undefined;
+  setAcousticModels(value?: AcousticModels): S2TInference;
+  hasAcousticModels(): boolean;
+  clearAcousticModels(): S2TInference;
 
   getLanguageModels(): LanguageModels | undefined;
   setLanguageModels(value?: LanguageModels): S2TInference;
   hasLanguageModels(): boolean;
   clearLanguageModels(): S2TInference;
 
+  getInferenceBackend(): InferenceBackend;
+  setInferenceBackend(value: InferenceBackend): S2TInference;
+
   serializeBinary(): Uint8Array;
   toObject(includeInstance?: boolean): S2TInference.AsObject;
   static toObject(includeInstance: boolean, msg: S2TInference): S2TInference.AsObject;
@@ -688,50 +703,119 @@ export class S2TInference extends jspb.Message {
 
 export namespace S2TInference {
   export type AsObject = {
-    ctcAcousticModels?: CtcAcousticModels.AsObject,
+    acousticModels?: AcousticModels.AsObject,
     languageModels?: LanguageModels.AsObject,
+    inferenceBackend: InferenceBackend,
   }
 }
 
-export class CtcAcousticModels extends jspb.Message {
+export class AcousticModels extends jspb.Message {
   getType(): string;
-  setType(value: string): CtcAcousticModels;
+  setType(value: string): AcousticModels;
 
   getQuartznet(): Quartznet | undefined;
-  setQuartznet(value?: Quartznet): CtcAcousticModels;
+  setQuartznet(value?: Quartznet): AcousticModels;
   hasQuartznet(): boolean;
-  clearQuartznet(): CtcAcousticModels;
+  clearQuartznet(): AcousticModels;
 
   getQuartznetTriton(): QuartznetTriton | undefined;
-  setQuartznetTriton(value?: QuartznetTriton): CtcAcousticModels;
+  setQuartznetTriton(value?: QuartznetTriton): AcousticModels;
   hasQuartznetTriton(): boolean;
-  clearQuartznetTriton(): CtcAcousticModels;
+  clearQuartznetTriton(): AcousticModels;
 
   getWav2vec(): Wav2Vec | undefined;
-  setWav2vec(value?: Wav2Vec): CtcAcousticModels;
+  setWav2vec(value?: Wav2Vec): AcousticModels;
   hasWav2vec(): boolean;
-  clearWav2vec(): CtcAcousticModels;
+  clearWav2vec(): AcousticModels;
 
   getWav2vecTriton(): Wav2VecTriton | undefined;
-  setWav2vecTriton(value?: Wav2VecTriton): CtcAcousticModels;
+  setWav2vecTriton(value?: Wav2VecTriton): AcousticModels;
   hasWav2vecTriton(): boolean;
-  clearWav2vecTriton(): CtcAcousticModels;
+  clearWav2vecTriton(): AcousticModels;
+
+  getWhisper(): Whisper | undefined;
+  setWhisper(value?: Whisper): AcousticModels;
+  hasWhisper(): boolean;
+  clearWhisper(): AcousticModels;
+
+  getWhisperTriton(): WhisperTriton | undefined;
+  setWhisperTriton(value?: WhisperTriton): AcousticModels;
+  hasWhisperTriton(): boolean;
+  clearWhisperTriton(): AcousticModels;
 
   serializeBinary(): Uint8Array;
-  toObject(includeInstance?: boolean): CtcAcousticModels.AsObject;
-  static toObject(includeInstance: boolean, msg: CtcAcousticModels): CtcAcousticModels.AsObject;
-  static serializeBinaryToWriter(message: CtcAcousticModels, writer: jspb.BinaryWriter): void;
-  static deserializeBinary(bytes: Uint8Array): CtcAcousticModels;
-  static deserializeBinaryFromReader(message: CtcAcousticModels, reader: jspb.BinaryReader): CtcAcousticModels;
+  toObject(includeInstance?: boolean): AcousticModels.AsObject;
+  static toObject(includeInstance: boolean, msg: AcousticModels): AcousticModels.AsObject;
+  static serializeBinaryToWriter(message: AcousticModels, writer: jspb.BinaryWriter): void;
+  static deserializeBinary(bytes: Uint8Array): AcousticModels;
+  static deserializeBinaryFromReader(message: AcousticModels, reader: jspb.BinaryReader): AcousticModels;
 }
 
-export namespace CtcAcousticModels {
+export namespace AcousticModels {
   export type AsObject = {
     type: string,
     quartznet?: Quartznet.AsObject,
     quartznetTriton?: QuartznetTriton.AsObject,
     wav2vec?: Wav2Vec.AsObject,
     wav2vecTriton?: Wav2VecTriton.AsObject,
+    whisper?: Whisper.AsObject,
+    whisperTriton?: WhisperTriton.AsObject,
+  }
+}
+
+export class Whisper extends jspb.Message {
+  getModelPath(): string;
+  setModelPath(value: string): Whisper;
+
+  getUseGpu(): boolean;
+  setUseGpu(value: boolean): Whisper;
+
+  getLanguage(): string;
+  setLanguage(value: string): Whisper;
+
+  serializeBinary(): Uint8Array;
+  toObject(includeInstance?: boolean): Whisper.AsObject;
+  static toObject(includeInstance: boolean, msg: Whisper): Whisper.AsObject;
+  static serializeBinaryToWriter(message: Whisper, writer: jspb.BinaryWriter): void;
+  static deserializeBinary(bytes: Uint8Array): Whisper;
+  static deserializeBinaryFromReader(message: Whisper, reader: jspb.BinaryReader): Whisper;
+}
+
+export namespace Whisper {
+  export type AsObject = {
+    modelPath: string,
+    useGpu: boolean,
+    language: string,
+  }
+}
+
+export class WhisperTriton extends jspb.Message {
+  getProcessorPath(): string;
+  setProcessorPath(value: string): WhisperTriton;
+
+  getTritonModelName(): string;
+  setTritonModelName(value: string): WhisperTriton;
+
+  getTritonModelVersion(): string;
+  setTritonModelVersion(value: string): WhisperTriton;
+
+  getCheckStatusTimeout(): number;
+  setCheckStatusTimeout(value: number): WhisperTriton;
+
+  serializeBinary(): Uint8Array;
+  toObject(includeInstance?: boolean): WhisperTriton.AsObject;
+  static toObject(includeInstance: boolean, msg: WhisperTriton): WhisperTriton.AsObject;
+  static serializeBinaryToWriter(message: WhisperTriton, writer: jspb.BinaryWriter): void;
+  static deserializeBinary(bytes: Uint8Array): WhisperTriton;
+  static deserializeBinaryFromReader(message: WhisperTriton, reader: jspb.BinaryReader): WhisperTriton;
+}
+
+export namespace WhisperTriton {
+  export type AsObject = {
+    processorPath: string,
+    tritonModelName: string,
+    tritonModelVersion: string,
+    checkStatusTimeout: number,
   }
 }
 
@@ -961,8 +1045,8 @@ export class StreamingSpeechRecognition extends jspb.Message {
   getTranscribeNotFinal(): boolean;
   setTranscribeNotFinal(value: boolean): StreamingSpeechRecognition;
 
-  getCtcDecodingMethod(): string;
-  setCtcDecodingMethod(value: string): StreamingSpeechRecognition;
+  getDecodingMethod(): string;
+  setDecodingMethod(value: string): StreamingSpeechRecognition;
 
   getSamplingRate(): number;
   setSamplingRate(value: number): StreamingSpeechRecognition;
@@ -990,7 +1074,7 @@ export class StreamingSpeechRecognition extends jspb.Message {
 export namespace StreamingSpeechRecognition {
   export type AsObject = {
     transcribeNotFinal: boolean,
-    ctcDecodingMethod: string,
+    decodingMethod: string,
     samplingRate: number,
     minAudioChunkSize: number,
     startOfUtteranceThreshold: number,
@@ -1372,8 +1456,14 @@ export namespace TrainUserLanguageModelRequest {
   }
 }
 
-export enum CTCDecoding { 
+export enum InferenceBackend { 
+  INFERENCE_BACKEND_UNKNOWN = 0,
+  INFERENCE_BACKEND_PYTORCH = 1,
+  INFERENCE_BACKEND_FLAX = 2,
+}
+export enum Decoding { 
   DEFAULT = 0,
   GREEDY = 1,
   BEAM_SEARCH_WITH_LM = 2,
+  BEAM_SEARCH = 3,
 }