initial work to remove speech to text controller from pangea controller

lib/pangea/common/controllers/pangea_controller.dart

@@ -21,8 +21,8 @@ import 'package:fluffychat/pangea/learning_settings/controllers/language_control
 import 'package:fluffychat/pangea/learning_settings/utils/locale_provider.dart';
 import 'package:fluffychat/pangea/learning_settings/utils/p_language_store.dart';
 import 'package:fluffychat/pangea/spaces/controllers/space_code_controller.dart';
+import 'package:fluffychat/pangea/speech_to_text/speech_to_text_controller.dart';
 import 'package:fluffychat/pangea/subscription/controllers/subscription_controller.dart';
-import 'package:fluffychat/pangea/toolbar/controllers/speech_to_text_controller.dart';
 import 'package:fluffychat/pangea/toolbar/controllers/text_to_speech_controller.dart';
 import 'package:fluffychat/pangea/toolbar/controllers/tts_controller.dart';
 import 'package:fluffychat/pangea/user/controllers/permissions_controller.dart';

lib/pangea/events/event_wrappers/pangea_message_event.dart

@@ -19,9 +19,10 @@ import 'package:fluffychat/pangea/events/repo/language_detection_request.dart';
 import 'package:fluffychat/pangea/events/repo/language_detection_response.dart';
 import 'package:fluffychat/pangea/learning_settings/utils/p_language_store.dart';
 import 'package:fluffychat/pangea/spaces/models/space_model.dart';
+import 'package:fluffychat/pangea/speech_to_text/audio_encoding_enum.dart';
+import 'package:fluffychat/pangea/speech_to_text/speech_to_text_request_model.dart';
+import 'package:fluffychat/pangea/speech_to_text/speech_to_text_response_model.dart';
 import 'package:fluffychat/pangea/toolbar/controllers/text_to_speech_controller.dart';
-import 'package:fluffychat/pangea/toolbar/enums/audio_encoding_enum.dart';
-import 'package:fluffychat/pangea/toolbar/models/speech_to_text_models.dart';
 import 'package:fluffychat/pangea/toolbar/widgets/message_audio_card.dart';
 import 'package:fluffychat/pangea/translation/full_text_translation_request_model.dart';
 import 'package:fluffychat/widgets/future_loading_dialog.dart';
@@ -227,13 +228,13 @@ class PangeaMessageEvent {
             null;
       }).toSet();
 
-  SpeechToTextModel? getSpeechToTextLocal() {
+  SpeechToTextResponseModel? getSpeechToTextLocal() {
     final rawBotTranscription =
         event.content.tryGetMap(ModelKey.botTranscription);
 
     if (rawBotTranscription != null) {
       try {
-        return SpeechToTextModel.fromJson(
+        return SpeechToTextResponseModel.fromJson(
           Map<String, dynamic>.from(rawBotTranscription),
         );
       } catch (err, s) {
@@ -257,7 +258,7 @@ class PangeaMessageEvent {
         .speechToText;
   }
 
-  Future<SpeechToTextModel> getSpeechToText(
+  Future<SpeechToTextResponseModel> getSpeechToText(
     String l1Code,
     String l2Code,
   ) async {
@@ -268,7 +269,8 @@ class PangeaMessageEvent {
     final rawBotTranscription =
         event.content.tryGetMap(ModelKey.botTranscription);
     if (rawBotTranscription != null) {
-      final SpeechToTextModel botTranscription = SpeechToTextModel.fromJson(
+      final SpeechToTextResponseModel botTranscription =
+          SpeechToTextResponseModel.fromJson(
         Map<String, dynamic>.from(rawBotTranscription),
       );
 
@@ -290,7 +292,7 @@ class PangeaMessageEvent {
       return botTranscription;
     }
 
-    final SpeechToTextModel? speechToTextLocal = representations
+    final SpeechToTextResponseModel? speechToTextLocal = representations
         .firstWhereOrNull(
           (element) => element.content.speechToText != null,
         )
@@ -303,7 +305,7 @@ class PangeaMessageEvent {
 
     final matrixFile = await _event.downloadAndDecryptAttachment();
 
-    final SpeechToTextModel response =
+    final SpeechToTextResponseModel response =
         await MatrixState.pangeaController.speechToText.get(
       SpeechToTextRequestModel(
         audioContent: matrixFile.bytes,

lib/pangea/events/models/representation_content_model.dart

@@ -8,7 +8,7 @@ import 'package:fluffychat/pangea/choreographer/choreo_record_model.dart';
 import 'package:fluffychat/pangea/choreographer/igc/pangea_match_status_enum.dart';
 import 'package:fluffychat/pangea/common/utils/error_handler.dart';
 import 'package:fluffychat/pangea/events/models/pangea_token_model.dart';
-import 'package:fluffychat/pangea/toolbar/models/speech_to_text_models.dart';
+import 'package:fluffychat/pangea/speech_to_text/speech_to_text_response_model.dart';
 import 'package:fluffychat/widgets/matrix.dart';
 
 /// this class is contained within a [RepresentationEvent]
@@ -30,7 +30,7 @@ class PangeaRepresentation {
   bool originalWritten;
 
   // a representation can be create via speech to text on the original message
-  SpeechToTextModel? speechToText;
+  SpeechToTextResponseModel? speechToText;
 
   // how do we know which representation was sent by author?
   // RepresentationEvent.text == PangeaMessageEvent.event.body
@@ -70,7 +70,7 @@ class PangeaRepresentation {
       originalWritten: json[_originalWrittenKey] ?? false,
       speechToText: json[_speechToTextKey] == null
           ? null
-          : SpeechToTextModel.fromJson(json[_speechToTextKey]),
+          : SpeechToTextResponseModel.fromJson(json[_speechToTextKey]),
     );
   }
 

lib/pangea/speech_to_text/audio_encoding_enum.dart

@@ -26,11 +26,8 @@ enum AudioEncodingEnum {
   speexWithHeaderByte,
   mp3,
   mp4,
-  webmOpus,
-}
+  webmOpus;
 
-// Utility extension to map enum values to their corresponding string value as used by the API
-extension AudioEncodingExtension on AudioEncodingEnum {
   String get value {
     switch (this) {
       case AudioEncodingEnum.linear16:

lib/pangea/speech_to_text/speech_to_text_repo.dart

@@ -0,0 +1,105 @@
+import 'dart:convert';
+
+import 'package:async/async.dart';
+import 'package:http/http.dart';
+
+import 'package:fluffychat/pangea/common/config/environment.dart';
+import 'package:fluffychat/pangea/common/network/requests.dart';
+import 'package:fluffychat/pangea/common/network/urls.dart';
+import 'package:fluffychat/pangea/common/utils/error_handler.dart';
+import 'package:fluffychat/pangea/speech_to_text/speech_to_text_request_model.dart';
+import 'package:fluffychat/pangea/speech_to_text/speech_to_text_response_model.dart';
+
+class _SpeechToTextCacheItem {
+  final Future<SpeechToTextResponseModel> data;
+  final DateTime timestamp;
+
+  const _SpeechToTextCacheItem({
+    required this.data,
+    required this.timestamp,
+  });
+}
+
+class SpeechToTextRepo {
+  static final Map<String, _SpeechToTextCacheItem> _cache = {};
+  static const Duration _cacheDuration = Duration(minutes: 10);
+
+  static Future<Result<SpeechToTextResponseModel>> get(
+    String accessToken,
+    SpeechToTextRequestModel request,
+  ) {
+    final cached = _getCached(request);
+    if (cached != null) {
+      return _getResult(request, cached);
+    }
+
+    final future = _fetch(accessToken, request);
+    _setCached(request, future);
+    return _getResult(request, future);
+  }
+
+  static Future<SpeechToTextResponseModel> _fetch(
+    String accessToken,
+    SpeechToTextRequestModel request,
+  ) async {
+    final Requests req = Requests(
+      choreoApiKey: Environment.choreoApiKey,
+      accessToken: accessToken,
+    );
+
+    final Response res = await req.post(
+      url: PApiUrls.simpleTranslation,
+      body: request.toJson(),
+    );
+
+    if (res.statusCode != 200) {
+      throw Exception(
+        'Failed to translate text: ${res.statusCode} ${res.reasonPhrase}',
+      );
+    }
+
+    return SpeechToTextResponseModel.fromJson(
+      jsonDecode(utf8.decode(res.bodyBytes)),
+    );
+  }
+
+  static Future<Result<SpeechToTextResponseModel>> _getResult(
+    SpeechToTextRequestModel request,
+    Future<SpeechToTextResponseModel> future,
+  ) async {
+    try {
+      final res = await future;
+      return Result.value(res);
+    } catch (e, s) {
+      _cache.remove(request.hashCode.toString());
+      ErrorHandler.logError(
+        e: e,
+        s: s,
+        data: request.toJson(),
+      );
+      return Result.error(e);
+    }
+  }
+
+  static Future<SpeechToTextResponseModel>? _getCached(
+    SpeechToTextRequestModel request,
+  ) {
+    final cacheKeys = [..._cache.keys];
+    for (final key in cacheKeys) {
+      if (DateTime.now().difference(_cache[key]!.timestamp) >= _cacheDuration) {
+        _cache.remove(key);
+      }
+    }
+
+    return _cache[request.hashCode.toString()]?.data;
+  }
+
+  static void _setCached(
+    SpeechToTextRequestModel request,
+    Future<SpeechToTextResponseModel> response,
+  ) =>
+      _cache[request.hashCode.toString()] = _SpeechToTextCacheItem(
+        data: response,
+        timestamp: DateTime.now(),
+      );
+}

lib/pangea/speech_to_text/speech_to_text_request_model.dart

@@ -0,0 +1,70 @@
+import 'dart:convert';
+
+import 'package:flutter/foundation.dart';
+
+import 'package:matrix/matrix.dart';
+
+import 'package:fluffychat/pangea/speech_to_text/audio_encoding_enum.dart';
+
+class SpeechToTextRequestModel {
+  final Uint8List audioContent;
+  final SpeechToTextAudioConfigModel config;
+  final Event? audioEvent;
+
+  SpeechToTextRequestModel({
+    required this.audioContent,
+    required this.config,
+    this.audioEvent,
+  });
+
+  Map<String, dynamic> toJson() => {
+        "audio_content": base64Encode(audioContent),
+        "config": config.toJson(),
+      };
+
+  @override
+  bool operator ==(Object other) {
+    if (identical(this, other)) return true;
+    if (other is! SpeechToTextRequestModel) return false;
+
+    return listEquals(audioContent, other.audioContent) &&
+        config == other.config;
+  }
+
+  @override
+  int get hashCode {
+    final bytesSample =
+        audioContent.length > 10 ? audioContent.sublist(0, 10) : audioContent;
+    return Object.hashAll([
+      Object.hashAll(bytesSample),
+      config.hashCode,
+    ]);
+  }
+}
+
+class SpeechToTextAudioConfigModel {
+  final AudioEncodingEnum encoding;
+  final int sampleRateHertz;
+  final bool enableWordConfidence;
+  final bool enableAutomaticPunctuation;
+  final String userL1;
+  final String userL2;
+
+  SpeechToTextAudioConfigModel({
+    required this.encoding,
+    required this.userL1,
+    required this.userL2,
+    this.sampleRateHertz = 16000,
+    this.enableWordConfidence = true,
+    this.enableAutomaticPunctuation = true,
+  });
+
+  Map<String, dynamic> toJson() => {
+        "encoding": encoding.value,
+        "sample_rate_hertz": sampleRateHertz,
+        "user_l1": userL1,
+        "user_l2": userL2,
+        "enable_word_confidence": enableWordConfidence,
+        "enable_automatic_punctuation": enableAutomaticPunctuation,
+      };
+}

lib/pangea/speech_to_text/speech_to_text_response_model.dart

@@ -1,79 +1,118 @@
-import 'dart:convert';
-
-import 'package:flutter/foundation.dart';
 import 'package:flutter/material.dart';
 
-import 'package:matrix/matrix.dart';
-
 import 'package:fluffychat/config/app_config.dart';
 import 'package:fluffychat/pangea/analytics_misc/construct_use_type_enum.dart';
 import 'package:fluffychat/pangea/analytics_misc/constructs_model.dart';
 import 'package:fluffychat/pangea/events/models/pangea_token_model.dart';
-import 'package:fluffychat/pangea/toolbar/enums/audio_encoding_enum.dart';
 
-const int thresholdForGreen = 80;
+class SpeechToTextResponseModel {
+  final List<SpeechToTextResult> results;
 
-class SpeechToTextAudioConfigModel {
-  final AudioEncodingEnum encoding;
-  final int sampleRateHertz;
-  final bool enableWordConfidence;
-  final bool enableAutomaticPunctuation;
-  final String userL1;
-  final String userL2;
-
-  SpeechToTextAudioConfigModel({
-    required this.encoding,
-    required this.userL1,
-    required this.userL2,
-    this.sampleRateHertz = 16000,
-    this.enableWordConfidence = true,
-    this.enableAutomaticPunctuation = true,
+  SpeechToTextResponseModel({
+    required this.results,
   });
 
+  Transcript get transcript => results.first.transcripts.first;
+
+  String get langCode => results.first.transcripts.first.langCode;
+
+  factory SpeechToTextResponseModel.fromJson(Map<String, dynamic> json) {
+    final results = json['results'] as List;
+    if (results.isEmpty) {
+      throw Exception('SpeechToTextModel.fromJson: results is empty');
+    }
+    return SpeechToTextResponseModel(
+      results: (json['results'] as List)
+          .map((e) => SpeechToTextResult.fromJson(e))
+          .toList(),
+    );
+  }
+
   Map<String, dynamic> toJson() => {
-        "encoding": encoding.value,
-        "sample_rate_hertz": sampleRateHertz,
-        "user_l1": userL1,
-        "user_l2": userL2,
-        "enable_word_confidence": enableWordConfidence,
-        "enable_automatic_punctuation": enableAutomaticPunctuation,
+        "results": results.map((e) => e.toJson()).toList(),
+      };
+
+  List<OneConstructUse> constructs(
+    String roomId,
+    String eventId,
+  ) {
+    final List<OneConstructUse> constructs = [];
+    final metadata = ConstructUseMetaData(
+      roomId: roomId,
+      eventId: eventId,
+      timeStamp: DateTime.now(),
+    );
+    for (final sstToken in transcript.sttTokens) {
+      final token = sstToken.token;
+      if (!token.lemma.saveVocab) continue;
+      constructs.addAll(
+        token.allUses(
+          ConstructUseTypeEnum.pvm,
+          metadata,
+          ConstructUseTypeEnum.pvm.pointValue,
+        ),
+      );
+    }
+    return constructs;
+  }
+}
+
+class SpeechToTextResult {
+  final List<Transcript> transcripts;
+
+  SpeechToTextResult({required this.transcripts});
+
+  factory SpeechToTextResult.fromJson(Map<String, dynamic> json) =>
+      SpeechToTextResult(
+        transcripts: (json['transcripts'] as List)
+            .map((e) => Transcript.fromJson(e))
+            .toList(),
+      );
+
+  Map<String, dynamic> toJson() => {
+        "transcripts": transcripts.map((e) => e.toJson()).toList(),
       };
 }
 
-class SpeechToTextRequestModel {
-  final Uint8List audioContent;
-  final SpeechToTextAudioConfigModel config;
-  final Event? audioEvent;
+class Transcript {
+  final String text;
+  final int confidence;
+  final List<STTToken> sttTokens;
+  final String langCode;
+  final int? wordsPerHr;
 
-  SpeechToTextRequestModel({
-    required this.audioContent,
-    required this.config,
-    this.audioEvent,
+  Transcript({
+    required this.text,
+    required this.confidence,
+    required this.sttTokens,
+    required this.langCode,
+    required this.wordsPerHr,
   });
 
+  /// Returns the number of words per minute rounded to one decimal place.
+  double? get wordsPerMinute => wordsPerHr != null ? wordsPerHr! / 60 : null;
+
+  factory Transcript.fromJson(Map<String, dynamic> json) => Transcript(
+        text: json['transcript'],
+        confidence: json['confidence'] <= 100
+            ? json['confidence']
+            : json['confidence'] / 100,
+        sttTokens: (json['stt_tokens'] as List)
+            .map((e) => STTToken.fromJson(e))
+            .toList(),
+        langCode: json['lang_code'],
+        wordsPerHr: json['words_per_hr'],
+      );
+
   Map<String, dynamic> toJson() => {
-        "audio_content": base64Encode(audioContent),
-        "config": config.toJson(),
+        "transcript": text,
+        "confidence": confidence,
+        "stt_tokens": sttTokens.map((e) => e.toJson()).toList(),
+        "lang_code": langCode,
+        "words_per_hr": wordsPerHr,
       };
 
-  @override
-  bool operator ==(Object other) {
-    if (identical(this, other)) return true;
-    if (other is! SpeechToTextRequestModel) return false;
-
-    return listEquals(audioContent, other.audioContent) &&
-        config == other.config;
-  }
-
-  @override
-  int get hashCode {
-    final bytesSample =
-        audioContent.length > 10 ? audioContent.sublist(0, 10) : audioContent;
-    return Object.hashAll([
-      Object.hashAll(bytesSample),
-      config.hashCode,
-    ]);
-  }
+  Color get color => confidence > 80 ? AppConfig.success : AppConfig.warning;
 }
 
 class STTToken {
@@ -94,15 +133,7 @@ class STTToken {
   int get length => token.text.length;
 
   Color color(BuildContext context) {
-    // turning off the color coding for now
-    // whisper doesn't include word-level confidence
-    // if (confidence == null) {
     return Theme.of(context).colorScheme.onSurface;
-    // }
-    // if (confidence! > thresholdForGreen) {
-    //   return AppConfig.success;
-    // }
-    // return AppConfig.warning;
   }
 
   factory STTToken.fromJson(Map<String, dynamic> json) {
@@ -147,118 +178,3 @@ class STTToken {
     ]);
   }
 }
-
-class Transcript {
-  final String text;
-  final int confidence;
-  final List<STTToken> sttTokens;
-  final String langCode;
-  final int? wordsPerHr;
-
-  Transcript({
-    required this.text,
-    required this.confidence,
-    required this.sttTokens,
-    required this.langCode,
-    required this.wordsPerHr,
-  });
-
-  /// Returns the number of words per minute rounded to one decimal place.
-  double? get wordsPerMinute => wordsPerHr != null ? wordsPerHr! / 60 : null;
-
-  factory Transcript.fromJson(Map<String, dynamic> json) => Transcript(
-        text: json['transcript'],
-        confidence: json['confidence'] <= 100
-            ? json['confidence']
-            : json['confidence'] / 100,
-        sttTokens: (json['stt_tokens'] as List)
-            .map((e) => STTToken.fromJson(e))
-            .toList(),
-        langCode: json['lang_code'],
-        wordsPerHr: json['words_per_hr'],
-      );
-
-  Map<String, dynamic> toJson() => {
-        "transcript": text,
-        "confidence": confidence,
-        "stt_tokens": sttTokens.map((e) => e.toJson()).toList(),
-        "lang_code": langCode,
-        "words_per_hr": wordsPerHr,
-      };
-
-  Color color(BuildContext context) {
-    if (confidence > thresholdForGreen) {
-      return AppConfig.success;
-    }
-    return AppConfig.warning;
-  }
-}
-
-class SpeechToTextResult {
-  final List<Transcript> transcripts;
-
-  SpeechToTextResult({required this.transcripts});
-
-  factory SpeechToTextResult.fromJson(Map<String, dynamic> json) =>
-      SpeechToTextResult(
-        transcripts: (json['transcripts'] as List)
-            .map((e) => Transcript.fromJson(e))
-            .toList(),
-      );
-
-  Map<String, dynamic> toJson() => {
-        "transcripts": transcripts.map((e) => e.toJson()).toList(),
-      };
-}
-
-class SpeechToTextModel {
-  final List<SpeechToTextResult> results;
-
-  SpeechToTextModel({
-    required this.results,
-  });
-
-  Transcript get transcript => results.first.transcripts.first;
-
-  String get langCode => results.first.transcripts.first.langCode;
-
-  factory SpeechToTextModel.fromJson(Map<String, dynamic> json) {
-    final results = json['results'] as List;
-    if (results.isEmpty) {
-      throw Exception('SpeechToTextModel.fromJson: results is empty');
-    }
-    return SpeechToTextModel(
-      results: (json['results'] as List)
-          .map((e) => SpeechToTextResult.fromJson(e))
-          .toList(),
-    );
-  }
-
-  Map<String, dynamic> toJson() => {
-        "results": results.map((e) => e.toJson()).toList(),
-      };
-
-  List<OneConstructUse> constructs(
-    String roomId,
-    String eventId,
-  ) {
-    final List<OneConstructUse> constructs = [];
-    final metadata = ConstructUseMetaData(
-      roomId: roomId,
-      eventId: eventId,
-      timeStamp: DateTime.now(),
-    );
-    for (final sstToken in transcript.sttTokens) {
-      final token = sstToken.token;
-      if (!token.lemma.saveVocab) continue;
-      constructs.addAll(
-        token.allUses(
-          ConstructUseTypeEnum.pvm,
-          metadata,
-          ConstructUseTypeEnum.pvm.pointValue,
-        ),
-      );
-    }
-    return constructs;
-  }
-}

lib/pangea/toolbar/controllers/speech_to_text_controller.dart

@@ -1,133 +0,0 @@
-import 'dart:async';
-import 'dart:convert';
-
-import 'package:flutter/foundation.dart';
-
-import 'package:http/http.dart';
-
-import 'package:fluffychat/pangea/common/controllers/pangea_controller.dart';
-import 'package:fluffychat/pangea/common/utils/error_handler.dart';
-import 'package:fluffychat/pangea/events/constants/pangea_event_types.dart';
-import 'package:fluffychat/pangea/events/models/representation_content_model.dart';
-import 'package:fluffychat/pangea/extensions/pangea_room_extension.dart';
-import 'package:fluffychat/pangea/toolbar/models/speech_to_text_models.dart';
-import '../../common/config/environment.dart';
-import '../../common/network/requests.dart';
-import '../../common/network/urls.dart';
-
-// Assuming SpeechToTextRequestModel, SpeechToTextModel and related models are already defined as in your provided code.
-
-class _SpeechToTextCacheItem {
-  Future<SpeechToTextModel> data;
-
-  _SpeechToTextCacheItem({required this.data});
-}
-
-class SpeechToTextController {
-  static final Map<int, _SpeechToTextCacheItem> _cache = {};
-  late final PangeaController _pangeaController;
-  Timer? _cacheClearTimer;
-
-  SpeechToTextController(this._pangeaController) {
-    _initializeCacheClearing();
-  }
-
-  void _initializeCacheClearing() {
-    const duration = Duration(minutes: 2);
-    _cacheClearTimer = Timer.periodic(duration, (Timer t) => _clearCache());
-  }
-
-  void _clearCache() {
-    _cache.clear();
-  }
-
-  void dispose() {
-    _cacheClearTimer?.cancel();
-  }
-
-  Future<SpeechToTextModel> get(
-    SpeechToTextRequestModel requestModel,
-  ) async {
-    final int cacheKey = requestModel.hashCode;
-
-    if (_cache.containsKey(cacheKey)) {
-      return _cache[cacheKey]!.data;
-    } else {
-      final Future<SpeechToTextModel> response = _fetchResponse(
-        accessToken: _pangeaController.userController.accessToken,
-        requestModel: requestModel,
-      );
-      _cache[cacheKey] = _SpeechToTextCacheItem(data: response);
-
-      return response;
-    }
-  }
-
-  Future<void> saveSpeechToTextAsRepresentationEvent(
-    SpeechToTextModel response,
-    SpeechToTextRequestModel requestModel,
-  ) {
-    if (requestModel.audioEvent == null) {
-      debugPrint(
-        'Audio event is null, case of giving speech to text before message sent, currently not implemented',
-      );
-      return Future.value(null);
-    }
-    debugPrint('Saving transcript as matrix event');
-
-    requestModel.audioEvent?.room
-        .sendPangeaEvent(
-          content: PangeaRepresentation(
-            langCode: response.langCode,
-            text: response.transcript.text,
-            originalSent: false,
-            originalWritten: false,
-            speechToText: response,
-          ).toJson(),
-          parentEventId: requestModel.audioEvent!.eventId,
-          type: PangeaEventTypes.representation,
-        )
-        .then(
-          (_) => debugPrint('Transcript saved as matrix event'),
-        );
-
-    return Future.value(null);
-  }
-
-  Future<SpeechToTextModel> _fetchResponse({
-    required String accessToken,
-    required SpeechToTextRequestModel requestModel,
-  }) async {
-    final Requests request = Requests(
-      choreoApiKey: Environment.choreoApiKey,
-      accessToken: accessToken,
-    );
-
-    final Response res = await request.post(
-      url: PApiUrls.speechToText,
-      body: requestModel.toJson(),
-    );
-
-    if (res.statusCode == 200) {
-      final Map<String, dynamic> json = jsonDecode(utf8.decode(res.bodyBytes));
-
-      final response = SpeechToTextModel.fromJson(json);
-
-      saveSpeechToTextAsRepresentationEvent(response, requestModel).onError(
-        (error, stackTrace) => ErrorHandler.logError(
-          e: error,
-          s: stackTrace,
-          data: {
-            "response": response.toJson(),
-            "requestModel": requestModel.toJson(),
-          },
-        ),
-      );
-
-      return response;
-    } else {
-      debugPrint('Error converting speech to text: ${res.body}');
-      throw Exception('Failed to convert speech to text');
-    }
-  }
-}

lib/pangea/toolbar/widgets/select_mode_controller.dart

@@ -9,17 +9,17 @@ import 'package:path_provider/path_provider.dart';
 import 'package:fluffychat/pangea/common/utils/async_state.dart';
 import 'package:fluffychat/pangea/events/event_wrappers/pangea_message_event.dart';
 import 'package:fluffychat/pangea/events/extensions/pangea_event_extension.dart';
-import 'package:fluffychat/pangea/toolbar/models/speech_to_text_models.dart';
+import 'package:fluffychat/pangea/speech_to_text/speech_to_text_response_model.dart';
 import 'package:fluffychat/pangea/toolbar/widgets/message_audio_card.dart';
 import 'package:fluffychat/pangea/toolbar/widgets/select_mode_buttons.dart';
 import 'package:fluffychat/widgets/matrix.dart';
 
-class _TranscriptionLoader extends AsyncLoader<SpeechToTextModel> {
+class _TranscriptionLoader extends AsyncLoader<SpeechToTextResponseModel> {
   final PangeaMessageEvent messageEvent;
   _TranscriptionLoader(this.messageEvent) : super();
 
   @override
-  Future<SpeechToTextModel> fetch() => messageEvent.getSpeechToText(
+  Future<SpeechToTextResponseModel> fetch() => messageEvent.getSpeechToText(
         MatrixState.pangeaController.languageController.userL1!.langCodeShort,
         MatrixState.pangeaController.languageController.userL2!.langCodeShort,
       );
@@ -127,7 +127,7 @@ class SelectModeController {
   ValueNotifier<AsyncState<String>> get translationState =>
       _translationLoader.state;
 
-  ValueNotifier<AsyncState<SpeechToTextModel>> get transcriptionState =>
+  ValueNotifier<AsyncState<SpeechToTextResponseModel>> get transcriptionState =>
       _transcriptLoader.state;
 
   ValueNotifier<AsyncState<String>> get speechTranslationState =>

lib/pangea/toolbar/widgets/stt_transcript_tokens.dart

@@ -2,10 +2,10 @@ import 'package:flutter/material.dart';
 
 import 'package:fluffychat/pangea/events/models/pangea_token_model.dart';
 import 'package:fluffychat/pangea/message_token_text/tokens_util.dart';
-import 'package:fluffychat/pangea/toolbar/models/speech_to_text_models.dart';
+import 'package:fluffychat/pangea/speech_to_text/speech_to_text_response_model.dart';
 
 class SttTranscriptTokens extends StatelessWidget {
-  final SpeechToTextModel model;
+  final SpeechToTextResponseModel model;
   final TextStyle? style;
 
   final void Function(PangeaToken)? onClick;